Full text of "Mount Sinai Journal of Medicine"

See other formats

•J

Digitized by the Internet Archive
in 2015

https://archive.org/details/mountsinaijourna6019moun

^^H^^l o> '"i r >t jn^ipiot

lOONT SINAI JOORNAL
)F MEDIQNE

LGME 60 NUMBER 1

JANUARY 1993

Issues in Medical Ethics

1. The Doctor-Patient Relationship
2. Transplantation: Ethics and Organ Procurement

Daniel A. Moros and Rosamond Rhodes, guest editors
From the Mount Sinai School of Medicine (CUNY)

plus

David Z. Starr on a doctor's experience as a hospital patient
Kenneth I. Shine on science and responsibility
Faye Wattleton on the political responsibility of doctors

Beth

Veterans Israel

Affairs Medical

Center

¥1^ The Mount Sinai Journal of Medicine is published by The Mount Sinai Medical Center of
I 11 "^^^^ has the following affiliates: Beth Israel Medical Center, New York; Bronx

MOUNT SINAI
JOURNAL OF
MEDICINE

Veterans Affairs Medical Center, New York; and Elnnhurst
Hospital Center, New York.

Editor-in-Chief

Sherman Kupfer, M.D.

Editor Emeritus

Lester R. Tuchman, M.D.

Associate Editors

Harriet S. Gilbert, M.D. Julius Wolf, M.D.

Managing Editor

Claire Sotnick

Business and Production Assistant

Karen Schwartz

Assistant Editors

Stephen G. Baum, M.D.
David H. Bechhofer, Ph.D.
Constanin A. Bona, M.D.. Ph.D.
Edward J. Bottone, Ph.D.
Jurgen Brosius, Ph.D.
Lewis Burrows, M.D.
Joseph S. Eisenman, Ph.D.
Adrienne M. Fleckman. M.D.
Richard A. Frieden. M.D.
Steven Fruchtman, M.D.
Paul L. Gilbert. M.D.
James H. Godbold, Ph.D.

Richard S. Haber. M.D.
Noam Harpaz, M.D.
Dennis P. Healy, Ph.D.
Tomas Heimann, M.D.
Barry W. Jaffin, M.D.
Andrew S. Kaplan, D.D.S.
Samuel Kenan, M.D.
Suzanne Carter Kramer, M.
Mark G. Lebwohl, M.D.
Kenneth Lieberman, M.D.
Charles Lockwood, M.D.

Steven Markowitz, M.D.
Bernard Mehl, D.P.S.
Myron Miller, M.D.
Edward Raab. M.D.
Allan Reed, M.D.
Allan E. Rubenstein, M.D.
David B. Sachar, M.D.
Henry Sacks, M.D.
Robert Safirstein, M.D.
Ira Sanders, M.D.
Martin H. Savitz, M.D.

Clyde B. Schechter, M.D.
Michael Serby, M.D.
Phyllis Shaw, Ph.D.
George Silvay, M.D.
Barry D. Stimmel, M.D.
Nelson Stone, M.D.
Max Sung, M.D.
Carl Teplitz, M.D.
Rein Tideiksaar, Ph.D.
Richard P. Wedeen, M.D.
Michael F. Wesson, M.D.

Editorial Board

Barry Freedman, M.B.A.
Richard Gorlin. M.D.
Nathan Kase, M.D.

Panayotis G. Katsoyannis, Ph.D.
Charles K. McSherry, M.D.
Jack G. Rabinowitz, M.D.

John W. Rowe, M.D.
Alan L. Schiller. M.D.
Alan L. Silver, M.D.

Alvin S. Teirstein, M.D.
Rosalyn S. Yalow, Ph.D.

The Mount Sinai Journal of Medicine (ISSN No. 0027-2507; USPS 284-860) is published 6 times a year in January, March. May,
September, October, and November in one indexed volume by the Committee on Medical Education and Publications (Owner), The
Mount Sinai Hospital, Box 1094, One Gustave L. Levy Place, New York, NY 10029-6574. Subscription price: individuals. U.S.,
$65 per year; libraries, $70; individuals outside the U.S., $75. Single copies. $15. New subscriptions begin with the first issue of
the calendar year. Send notice of change of address 60 days before the change is effective. Second-class postage paid at New York,
NY and at additional mailing offices. POSTMASTER: Send address changes to The Mount Sinai Journal of Medicine, Box 1094, One
Gustave L. Levy Place. New York, NY 10029-6574. Copyright 1993 The Mount Sinai Hospital.

Statement of Ownership, Management, and Circulation filed with the U.S. Postal Service. Average no. copies each issue during
preceding 12 months: 10a. net press run. 2572; lObl. single copy sales, 53; 10b2. mail subscriptions, 2280; 10c. total paid or requested
circulation, 2314; lOd. nonpaid distribution. 200; lOe. total distribution, 2514; lOfl. copies not distributed, 58; 10f2. return from news
agents, none; lOg. total number of copies, 2572. Actual no. of copies of issue published nearest to filing date: 10a. net press run, 241 1;
lObl . single copy sales, none; 10b2. mail subscriptions, 2228; 10c. total paid or requested circulation, 2228; lOd. nonpaid distribution,
1(X); lOe. total distribution, 2328; lOfL copies not distributed, 83; 10f2. return from news agents, none; lOg. total number of copies, 241 L

THE

MOUNT SINAI JOURNAL «°

OF MEDICINE January 1993

CONTENTS

ISSUES IN MEDICAL ETHICS

Daniel A. Moros and Rosamond Rhodes, editors

Introduction Daniel A. Moros and Rosamond Rhodes 1

1. THE DOCTOR-PATIENT RELATIONSHIP
WHOSE CONTRACT IS IT?

Whose Contract and What Is It? The Doctor-Patient Agreement

Bernard Baumrin 3

The Contract Model of the Doctor-Patient Relationship: A Cri-
tique and an Alternative Ethics of Responsibility John
Ladd 6

The Physician's Virtues and Legitimate Self-interest in the Pa-
tient-Physician Contract Laurence B. McCullough 11

Consumerism Rampant: A Critique of the View of Medicine as a

Commercial Enterprise Daniel A. Moros 15

The Necessity and the Limitations of the Contract Model W. D.

White 20

THE PHYSICIAN'S EXPERIENCE

The Physician's Experience: Cases and Doubts Alisan B.
Goldfarb, Thomas H. Kalb, Bennett P. Leifer, and Audrey
Schwersenz 25

The Doctor-Patient Relationship: Panel Discussion Moderator:

Rosamond Rhodes 31

2. TRANSPLANTATION: ETHICS AND ORGAN PROCUREMENT
QUESTIONABLE DONORS

The Organ-Tissue Donation Process Angelina Korsun 37

Infants and Others Who Cannot Consent to Donation Thomas

Tomlinson 41

continued

Volume 60
Number 1
January 1993

CONTENTS continued

Debatable Donors: When Can We Count Their Consent? Rosa-
mond Rhodes 45

THE PHYSICIAN'S EXPERIENCE

The Physician's Experience: Cases and Doubts Steven M.

Fruchtman, Harry Schanzer, and Myron E. Schwartz 51

Cases and Doubts: Panel Discussion Moderator: Daniel A. Mo-

ros 55

GETTING AND GIVING

Those Who Don't Give Dena S. Davis 59

Legal Aspects of the Sale of Organs Beth Essig 64

Markets and Morals: The Case for Organ Sales Gerald Dwor-

kin 66

Getting and Giving: Panel Discussion Moderator: Rosamond

Rhodes 70

GENERAL ARTICLES

A Hospital Experience David Z. Starr 76

Equality, Justice, and Liberty: America's Unfinished Agenda

Faye Wattleton 79

Science, Scientists, and Responsibility Kenneth I. Shine 81

Poem: Heroism Jamel Oeser 84

Index to Advertisers

Critical Care America

page 85

Eli Lilly and Company

page 86

Introduction

Daniel A. Moros
Rosamond Rhodes

Issues in

Medical

Ethics

Concerns about our ability as a society to fairly and intelligently
distribute the increasingly expensive fruits of our technology, and
concerns about patient rights, patient autonomy, physician duties,
and the limits of medical paternalism have acquired increased
prominence as this century enters its final decade. Our ability to
intervene in human disease leads us to invade both body and
psyche in ways that earlier in this century would have seemed
science fiction. Yet the relationship of physician and patient re-
quires further understanding and explicit definition. In an earlier
era, when medicine offered little, the ethical issues embedded in
medical care, while interesting, only occasionally received serious
attention. But medical care and the health care budget can no
longer be treated casually. The need for individuals from many
disciplines — philosophy, law, government, as well as physicians —
to meet and discuss such issues is now well appreciated. This was
not so clearly the case in 1986, when the first in what has become
a yearly New York Regional Conference in Medical Ethics was
held at Mount Sinai Medical Center.

This issue of the Journal presents the proceedings of the 1991
and 1992 conferences sponsored by The Mount Sinai School of Med-
icine and The Ph.D. Program in Philosophy of The Graduate
School, CUNY. The proceedings of the 1987-1990 conferences have
been published in two earlier special issues of The Mount Sinai
Journal of Medicine. At the 1991 and 1992 conferences, practicing
physicians presented selected cases to challenge the philosophical
perspectives offered by the primary speakers. Each session ended
with a panel discussion between all participants.

The Doctor-Patient Relationship: Whose Contract Is It? Our
view of the proper conduct of the physician-patient relationship
has changed dramatically over the past three decades and reflects
our society's efforts to recognize individuals' rights for self-deter-
mination and to empower individuals to exercise their rights. This
change is most evident in the demand for informed consent and the
expectation that doctors will seriously try to understand the pa-
tient's objectives rather than simply paternalistically direct and
dictate the course of medical care.

But the doctor-patient relationship also has meaning for the
physician. It is not simply that the physician is entitled to earn
a living practicing medicine. Many physicians have chosen medi-
cine as a way of life and are deeply concerned with acting forcefully
and correctly. At the 1991 conference, philosophers Bernard
Baumrin, John Ladd, Laurence McCullough, and W. D. White and
physician Daniel A. Moros presented their views on the nature of
the doctor-patient relationship. Four young physicians, Alisan
Goldfarb, Tom Kalb, Bennett P. Leifer, and Audrey Schwersenz,
then presented cases from their practices which have raised doubts
about what the patient-physician relationship would require of
them.

For a panel discussion they were joined by Dr. Arthur H. Auf-
ses, Jr., chairman of Mount Sinai's Department of Surgery, Dr.
Richard Gorlin, chairman of Mount Sinai's Department of Medi-
cine, and Mount Sinai Director of Patient Representatives Ruth

Daniel A. Moros is Associate Professor of Neurology, Mount Sinai School of Med-
icine. Rosamond Rhodes is Assistant Professor of Medical Education and Director of
Bioethics Education, Mount Sinai School of Medicine.

The Mount Sinai Journal ok Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

Ravich. Among the issues considered in the 1991
conference are:

• Can this relationship be considered a straight-
forward commercial contract for service?

• What are the duties of a physician?

• Does the patient have any duties in the rela-
tionship?

• Are there special virtues that characterize the
good physician?

• Can the good physician act from "legitimate
self-interest"?

Transplantation: Ethics and Organ Procure-
ment. The ethical issues raised by organ trans-
plantation, where the recipient may be in desper-
ate need of a limited resource, where the living
donor may assume some risk, and where there
may be opposition to the use of cadaveric and fetal
tissue, illustrates how moral reasoning is chal-
lenged by our increasingly powerful ability to in-
tervene in human illness. The transplantation is-
sues are so various and complex that our 1992
conference focused entirely on the ethics of organ
procurement.

Angelina Korsun began by explaining the or-
gan procurement process. Philosophers Thomas
Tomlinson and Rosamond Rhodes then presented
their views on the acceptability of taking organs
from questionable donors. Medical transplant
specialists Steven Fruchtman, Harry Schanzer,
and Myron E. Schwartz then described cases from
their own experience with donors who did not
fully have the capacity to consent. All these par-
ticipants joined in a debate about the proper ap-

proach to the questionable donors whose cases
had been presented.

Getting and giving organs was the second fo-
cus of the 1992 conference. Attorney Beth Essig
explained the laws governing organ sales, and at-
torney Dena S. Davis and philosopher Gerald
Dworkin addressed philosophical issues raised by
the mechanics of getting organs. For discussion
they were joined by physicians Robert M. Arnold
from the Center for Medical Ethics at the Univer-
sity of Pittsburgh, Lewis Burrows, Mount Sinai's
director of kidney transplantation, and Charles
Miller, Mount Sinai's director of liver transplan-
tation. Among the issues considered in the pro-
ceedings are:

• May infants and others who lack the capacity
to consent be used as organ donors?

• Who cannot be a donor?

• Is it proper to sell organs?

• May a donor specify the recipient of a cadaveric
organ?

• Should those who do not promise to give organs
be eligible to receive them?

In this collection we again draw attention to
serious issues which are urgently confronting the
medical community. We remain committed to our
goal of providing a multidisciplinary forum for
wrestling with the emerging ethical issues which
confront medicine, so that those from the ivory
towers of academia and those from the trenches
who have to make painful medical decisions can
teach and learn from one another.

1. The Doctor 'Patient
Relationship

Whose Contract and What Is It?

The Doctor-Patient Agreement

Bernard Baumrin, Ph.D., J.D.

Most contracts are implicit; some others are
plain promises; only a few are written, and those
are created mainly to involve the courts (and the
police power the courts command) in enforce-
ment. The courts, of course, will also enforce some
oral promises, and they will enforce some implicit
contracts. I want to focus my attention on implied
contracts (and only derivatively on their enforce-
ment) because I see the contracts that exist in
medical care and service to be largely of this na-
ture. I say "largely" because to the extent that
they occur at all, oral promises and written agree-
ments merely punctuate the periphery of the im-
plicit core.

Implied Contracts. First, some examples of
simple implicit contracts. I go to the supermarket;
I pick up a can of tuna fish; I go to the checkout
person, who rings up the price, and I hand her the
money; she bags the tuna fish and I exit. All is
well; but to illustrate the contractual character of
this example, consider the sorts of things that
could be amiss. I go to the store; they do not let me
in, though they are open for business; I enter;
there is no tuna fish for sale (they have it, but not
for me, or only for special customers); there is
tuna fish, it has no posted price, and the price is
then fixed for me (in ways that I think nonstand-
ard, too high, too low); I go to the checkout person,
who rings up a different price; I go to the checkout
counter and walk through without paying; I pay,
she does not give me the tuna fish. All of these
events are breaches of contract; all of them, in
fact, are actionable in one way or the other, with

Adapted from the author's presentation at the conference
"Whose Contract Is It Anyway? Examining the Doctor-Patient
Relationship" at the Mount Sinai Medical Center on March 8,
1991. From the Department of Philosophy, Lehman College
and The Graduate School, CUNY. Address reprint requests to
the author at 590 West End Avenue, New York, NY 10024.

enforceable remedies, including injunctions, dam-
ages. And that, of course, is not the end of it. The
tuna fish may be contaminated; there may be less
tuna in the can than as labeled; there may be no
tuna in the can; the can of tuna may be part of a
catch of stolen tuna. The list is not endless, but it
is long.

A second example, having nothing to do with
the simple purchase of anything: I stop at a shoe-
shine stand, I sit, I put my feet up, and my shoes
are sandpapered, or changed from brown to black,
or even just ignored. In both of these examples, no
words need be spoken, yet it is perfectly clear that
the contracts exist, and it is perfectly clear what
sorts of thing would be a breach of the contract.

One of the deepest elements in contract law is
the concept of reliance. We can generally agree
that at least one party thought they had a con-
tract if they can be perceived as having relied on
(to their possible detriment) the behavior of the
other. So, to the extent that one party's behavior
tends to underlie a well-founded belief by the
other party that the first will do this or that, and
the believer acts as if he had the belief, the party
whose behavior gave rise to that belief is bound
contractually to satisfy the other's expectations.
His failure to do so is an actionable breach of con-
tract even if nothing has been said or written. Of
course, if something has been either said or writ-
ten, it will be easier to prove that the other's re-
liance was well-founded, but that is a mere mat-
ter of proof, not principle.

Estoppel. Along these lines, we have in law
an evidentiary doctrine which forestalls someone
from testifying that they did not do, or did not
intend to do, or did not say, or did not write what-
ever the complainer (plaintiff) relied on. This is
called an estoppel. In general, an estoppel oper-
ates when there is good evidence that someone
did, or said, or wrote, something; he or she is es-

The MouNfT Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

topped from denying it {merely because it would
now be helpful to do so). If the jury could form the
belief that the plaintiff claims to have formed, the
defendant will not be allowed to testify that he
did not do it, say it, write it, or mean it. He never
even gets to say the plaintiff is a fool, or gullible,
or even just plain mistaken.

Physicians and Patients. The foregoing is a
short version of a few elementary points about
contract. Many other critical points will come up
along the way. The person (we may in general
look at persons as patients from here on) is a pur-
chaser of services (like one who wishes his shoes
shined or craves a can of tuna fish). The physician
is an offeror. If this were all there was to it, then
the patient purchases what he wants, the physi-
cian sells it, and that would be the end of the
matter. It would not in principle differ from buy-
ing tuna fish or a shoe shine. But it is not the end
of the matter, for several significant reasons.
First, there is the "it": the patient wants health,
or life, or recovery of function, or repairs. The
larger the aim, the less likely it can easily be seen
to be in stock. If we went to the market for nour-
ishment, our quest would not be over because the
shelves are stocked with edibles, and if we went to
the shoe-shiner to look good, our shined shoes
would be no guarantor of fulfillment of our
search. Unshined shoes might not be our sartorial
or cosmetic problem.

The State as Third-Party Guarantor. In com-
ing to the physician, we go to one who is schooled
in designing aid for many types of quests, though
not all. He or she may have nothing for us, or
there just may not be anything for us. So to begin
with, the offeror (the doctor) is only offering a
limited stock, which is generally speaking an im-
plicit stock, so that the offeree (the patient) does
not know exactly what is for sale or what he or
she is getting. The patient may know that he is
getting a private hearing and some other well-
known aspects of the medical tradition, but of
particular services he is largely ignorant. A par-
tial solution to this problem is a state recognition
of the offeror's expertise. The state licenses the
offeror to sell an inchoate collection of wares, and
further bars others from purporting to sell them.
So into our doctor-patient relationship has
stepped a third-party guarantor that has made
itself a party to every such contract. The license
the physician gets implicitly assures buyers of the
wholesomeness of the goods.

The Doctoring Profession as Third-Party
Guarantor. We are, however, not at an end here,
for states are blind and notoriously ignorant.
They like to make laws but not necessarily know

any of the relevant facts. For this they turn to
experts, and here they turn to physicians (who
keep and have kept the arcane knowledge closely
held) and say, "You tell us who to license and we
will, but you had better not be mistaken or we
will throw the business open to everybody." So
the physicians' guild says, "O.K., we will tell you
who, and we will do that by training them. So if
we train them they count, and if we do not, they
do not." That done, the whole profession (or at
least those who train licensees) becomes another
third-party guarantor of each physician-patient
contract. By analogy, the profession assures qual-
ity control of the product, to which the state then
grants an exclusive distributorship.

Now we can easily see that the patient ap-
proaches a twice-guaranteed purveyor who has
been certified as an able selector and provider of
the relevant product. The doctoring profession
and the state have guaranteed that the patient
can rely on the quality of the goods offered, on
their being supplied in his interest, and on their
being in some sense wholesome. One might say
that the patient can expect that, cost aside, noth-
ing will be done that will not be in his best inter-
est, assuming of course that he has come for what
is in the doctor's bag of tricks. If it is not, he will
expect to be turned away with a polite but firm
"I'm sorry." Of course, each purveyor may have a
limited bag of tricks, so that the patient may have
to go to another purveyor. But in this the physi-
cian is duty-bound, that is, contractually bound,
to aid him or to pass him on to the right purveyor.

The physician for his or her part should be
operating with the confidence to which being li-
censed attests. He should believe that he has the
skills it attests to and that he is bound to provide
them on request, cost aside, and bound as well to
desist and refer to another if he does not have
those skills. He or she has prior contracts with the
profession and the state to act within the bounds
of his license, his professional imprimatur. He
may not morally violate that contract, for it is
implicit in the offer he makes to every patient. By
holding himself out as one who provides medical
care, he signals to the world that he knows what
he is doing and that he is guided by the standards
of the guild that successfully bars others from
holding themselves out as healers of this sort.
That is why confidentiality, as well as all the
other well-known traditions of the profession, can
be assumed.

Hospitals as Third-Party Guarantors. But so
far in this discussion another important party has
been ignored. Because of the complexity of mod-
ern medicine, it turns out that much that can be

Vol. 60 No. 1

DOCTOR-PATIENT CONTRACT— BAUMRIN

done must be done in larger institutions, partic-
ularly hospitals. The physician may not, because
of the standard of care now generally acknowl-
edged, do without the hospital what in former
times he might have done. He is barred from the
free-wheeling practice of medicine. And the insti-
tutions themselves (for many of the same reasons
as are cited above) became another group of third-
party guarantors of the physician-patient con-
tract. Hence, implicitly, when the patient seeks
medical help, he believes, and he is entitled to
believe, that the physician's aid is guaranteed,
certified, warranted to be wholesome and, to the
extent possible under the circumstances, per-
formed in his interest, for his ultimate good. So, of
course, there really is no reason for the patient to
overly concern himself or herself with the partic-
ulars of care.

The Patient's Knowledge of Quality of Care.
If it were not for the suspicion that not all physi-
cians are equally competent and not all physi-
cians see themselves as equally bound by the cri-
teria and rules of quality medical care, the
patient would be as entitled as the opener of a can
of tuna fish to use the service just as the prac-
tioner provides it and not as the purchaser
chooses or might wish to use it. The patient is not,
and generally cannot be, in a position to know
what he is getting, what he should be getting,
what is the best, what is adequate, what is sub-
standard. Such knowledge (from the patient's
point-of-view) is esoteric, beyond his ken, and he
relies and is entitled to rely on what is held out to
him. And the doctrine of estoppels holds the phy-
sician is not entitled after the fact to say he did
not know or did not imply he was able, or did not
mean to say that his actions would be in the pa-
tient's best interest. Nor can a hospital be heard
to say that it gave less than the best because it
was not able, did not promise, had no contract, did
not think the patient would rely on it.

Resistance to Care. In all of the above I have
left cost out because all the patient promises to do
is pay the bill and aid in his own care, or at least
not thwart the care extended. If he does not pay or
he resists being cared for, then he, for his part,
breaches the contract he has with the physician,
the hospital, and the profession. He, of course, can
be made to pay by the courts, but they have only
a limited ability to force the patient to assist in
his own care. Sometimes the court will step in and
force the patient to take the optimal path, but
often not. The patient's autonomy provides a
ground for him to break off the contract at nearly
every point. But, if he does, the responsibility for
the consequences is all his. If he does not break

the contract, the responsibility for any avoidable
bad consequences belongs to the physician, or the
hospital, or the profession (which licensed an in-
competent), or perhaps sometimes even the state
for licensing those that it ought not to have li-
censed.

The Quality of Third-Party Guarantees. I am

assuming here that there is something to doctor-
ing. If the state also licensed astrologers, I would
begin to have my doubts about the value of the
state's guarantee. Of course I do have my doubts
about the profession's guarantee. I know from law
and education that many are licensed who are
incompetent or become incompetent. Professions
notoriously do a bad job in policing themselves;
they should be held to a higher standard, and to
do so we should hold them collectively responsible
for the failings of their fellow artisans. We do
that, to some extent, through the high cost of mal-
practice insurance, penalizing all for the faults of
a few. In law we do it for criminal theft by lawyers
by providing a fund to compensate victims that is
paid for by all lawyers (at least in New York
State). Neither is quite enough, for what we need
is a more constant monitoring of the care pro-
vided. Physician review both by states and insur-
ance companies also aids in this endeavor, so I do
not mean to say there is no policing; but it is not
remarkably effective, and it is also not well
known. Were it effective and well known, the pa-
tient would be fully entitled to rely on the care
extended and to trust that there is good whole-
some tuna fish in every can.

The Implicit Contract between Physician
and Patient. Some will say that the picture I
paint is unrealistic. Patients differ, diseases dif-
fer, much is not known, each plays an enormous
role. All of that is true. Patients are not guaran-
teed health, or recovery, or youth, or beauty.
What they are guaranteed is the health provider's
best efforts, a level of concern directed to his in-
terest, and a level of quality that is informed,
able, and relevant to his case. Less than this is a
breach of the implied contract between them, just
as a breach of the implied duty of confidentiality
would be. No one is compelled to be a health pro-
vider. If anyone is not willing or not able there is
much else in the world to do. One could become a
philosopher; the pay is not so good, but the duties
to others are fewer and, after all, one does not
expect much of them. From physicians we expect
a great deal; we place our lives at their disposal,
and we have a well-founded belief that they will
take them as seriously as their own lives — that is
part of our (society's) implied contract with med-
icine.

The Contract Model of the
Doctor-Patient Relationship

A Critique and an Alternative Ethics of Responsibility

John Ladd, Ph.D.

My thesis is that the contractual model of the
doctor-patient relationship, when taken as an
ethical model, is incoherent, inappropriate, and
misleading, and is therefore morally unaccept-
able (1-4).

Since there is no single, crystal-clear, author-
itative version of the contract model of the doctor-
patient relationship for me to focus on in my at-
tack on the ethical adequacy of that model, I shall
have to proceed indirectly. Accordingly, my strat-
egy will be to review a number of typical and
ethically relevant features of contractualism in
general, in the hope thereby of being able to iso-
late facets of the contract notion that are or might
be relevant to an understanding and critical eval-
uation of the contractual model as applied to the
doctor-patient relationship. Although some of the
contentions in my analysis might not appear to be
germane, they are intended to be provocative. In
particular, they are intended as a challenge to
those who want to defend the contractualist posi-
tion to explain why the consequences and impli-
cations that I attribute to the contract notion in
general do not in fact pertain to the contract
model of the doctor-patient relationship.

Two Methodological Caveats. Before turning
to the substantive part of my discussion, I want to
mention two caveats, two traps to be avoided:

The first trap is the choice of an ethics by

default, as it were, that is, the assumption that
one has to choose between a limited set of ethical
options, for example, between paternalism and
contractualism or between a consequentialist and
a deontological ethics. I call this the "fallacy of
limited options." In fact, there are many other
available theories about the doctor-patient rela-
tionship, and if none of them is found to be satis-
factory, common sense would dictate that we
ought to start looking for new ones.

The second trap is mixing up ethics and law.
The confusion of the two leads to looking to the
law for the answers to ethical questions. This trap
might be called the "fallacy of the legalization of
ethics," the fallacy of deriving ethical conclusions
from legal premises alone.

In the rest of this essay, I shall assume that
the issue under discussion, contractualism, is an
ethical, not a legal issue.

Background. A quick look at the historical
background of contractualism may help to high-
light some of the essential facets of the theory.

The obvious historical antecedent to contrac-
tualism of the kind under consideration here is
the theory of social contract, as represented by
Hobbes, Locke, and others. The basic motivation
of the social contract theory was to develop an
answer to authoritarianism, particularly to the
notion of the divine right of kings.

In its best forms, the social contract theory is
antiauthoritarian, individualistic, and egalitar-
ian. And these are the obvious features of the con-
tract idea that those in quest of a model of the
doctor-patient relationship find attractive. Thus,
contractualism might be viewed as a modern re-
sponse to the exaggerated claims of medical pa-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

Vol. 60 No. 1

CRITIQUE OF THE CONTRACT MODEI^LADD

ternalism that suppose that doctors have a sort of
God-given right to rule over their patients.

There are other aspects of the social contract
theory, however, that are more problematic, such
as questions about when, how, and why the orig-
inal contract was made. Historically, in order to
save the theory, an extensive mythology had to be
invented to make things come out right. Thus, for
example, rather than being actual historical
events, the contracts were held to be fictitious,
tacit, implicit, or hypothetical. The same old
questions about the origins and basis of the con-
tract have come down to modern versions of con-
tractualism, including the contractualist model of
the doctor-patient relationship. We find again the
same need for a mythology to rescue the basic
idea.

(Contractualism, or "contractarianism," as it
is now called, has had a revival in contemporary
moral philosophy. It provides the cornerstone in
the otherwise diverse theories of philosophers
like Rawls, Gauthier, Scanlon, and others. The
motivation of these philosophers for reviving con-
tractarianism is, however, somewhat different
from that of the earlier historical theorists. Their
aim is more theoretical and is largely directed at
framing alternatives, as for example to utilitari-
anism and intuitionism.

For present purposes, perhaps the chief point
about all these different forms of contractualism,
old and new, is that they share the basic assump-
tion that the parties to the original contract are to
be conceived as self-interested individuals who
are "mutually unconcerned (do not care how oth-
ers fare)." Other-regarding interests and concerns
are automatically excluded in the original forma-
tion of the contract. I shall return to this point
later.

1 Questions. Bearing in mind that we are con-
1 cerned here with what might be called "moral
' contracts," rather than legal contracts, there are
still a number of questions that need to be asked
about such contracts, actual and alleged. I shall
simply mention but not try to answer any of these
questions here.

Who are the parties to the alleged contract?
How is it created: is it historical, fictitious, hypo-
thetical, constructive, tacit, implicit, and so on?
What kind of contract is it? What is it a contract
for? What is promised? What is the quid pro quo
for each party? What are the goods received and
exchanged by the parties? When is the contract
fulfilled, terminated, and so on? What are the
sanctions for violating the contract? And last but
not least: how does the contract affect the obliga-

tions and rights of third parties? I shall leave all
these questions to others to answer.

Ethical Aspects of Contracts
In General

To begin with, contracts are social mecha-
nisms for creating new rights and obligations,
over and above any other rights and duties that
individuals may already possess. The contract
model is situationally specific and specific to the
individuals involved, who, as just mentioned,
generally have no prior rights or obligations to
each other with regard to the matter at hand. For
example, in contracting to have my house
painted, the painter, who has no prior obligation
to me, acquires the new obligation to paint my
house and I, who have no prior obligation to the
painter, acquire the new obligation to pay him
the amount agreed upon. Prior to the contract,
neither of us had these obligations.

The plausibility of the contractual analysis of
the doctor-patient relationship rests in large part
on the consideration that when this relationship
is established, new rights and obligations come
into being that did not exist before. Thus the ob-
ligation of a doctor to treat a patient is not a con-
sequence of a general duty to help others; rather
it is an obligation that first comes into being
when the relationship is established. Contractu-
alists claim that the mutual obligations of doctors
and patients come into being through the making
of a contract.

It should be noted, however, that because
contracts are agreements and establish new rela-
tionships, it does not follow that all agreements
are contracts or that all new relationships, and
their corresponding mutual rights and obliga-
tions, are created through contracts. This is an
obvious non sequitur.

Another general fact about contracts is worth
noting. Not only do contracts create rights and
obligations de novo, so to speak, but they gener-
ally also stipulate the terms of the obligations,
that is, what is to be given and what is to be
received by each of the parties. Thus, it is the
nature of contracts in general to create both the
basis and the content of the mutual rights and
obligations — at least in theory.

Just as there is a certain arbitrariness about
whether there will be a contract or not, what goes
into a contract and what is left out is also arbi-
trary, inasmuch as it depends on what is agreed
to by the contractees. Within the limits of the law,
custom, and general morality, there is nothing in

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

principle to bar the contractees in a medical con-
tract from providing for quite idiosyncratic re-
quirements. Whimsical contracts are, at least
theoretically, still contracts. Does the alleged doc-
tor-patient contract permit whimsical contracts?

One obvious answer to this last question is
that the terms of the contract are set by the pro-
fession or other social institutions. In other
words, professional standards determine what is
or is not to be expected in the doctor-patient re-
lationship. But this, in turn, suggests that the
contractual aspect is redundant.

A second side of the contract model, in this
case a presupposition, is the principle of freedom
of contract. This principle requires that the par-
ties to a contract enter the contract freely and
willingly, in that the prior bargaining is unen-
cumbered and the parties bargain as equals. Ef-
fectually, this means that either of the contract-
ees must be in a position to refuse the offer of a
contract made by the other.

Here we must ask: Is the required freedom
and equality possible in the doctor-patient rela-
tionship? Are patients really free to refuse, and do
they bargain with doctors as equals? Can the con-
tractual model be made to fit without introducing
ad hoc and mythological assumptions, such as the
presumption that the patient, who is in fact un-
conscious, can be taken to have agreed to the con-
tract because he would have done so had he been
conscious, or because any rational person would
agree to such a contract? But, I submit, that sort
of mythology destroys the basic idea of a contract
as a binding agreement entered into freely among
equals. I need not labor this obvious difficulty.

A third side of the idea of a contract is that it
brackets questions of motivation. Once the con-
tract has been made, the source and rationality of
the contractee's interests are excluded from con-
sideration. As long as they are not provided for in
the contract, the needs of a patient and the desire
of a doctor to help the patient are irrelevant to
what is considered to be an obligation or a right
stemming from the contract.

I shall return to this point about the exclu-
sion of motivation later.

The Ideology of Contractualism. Contractu-
alism, more specifically contractarianism, as an
ethical theory characteristically presupposes a
particular view of human nature, of human rela-
tions, and of society, which, following Gauthier, I
shall call its "ideology."

The ideology in question starts with an
atomic individualism of some sort or other that
maintains that society consists of self-interested
individuals. The actions of these self-interested

individuals are coordinated for mutual self-inter-
est through contracts that they make with each
other in order to improve their situation. In the
final analysis, self-interest is the only rationale
for self-interested individuals to enter a contract.

According to the contractualist ideology, the
social bonds (that is, binding moral relations) be-
tween individuals in society are contractual in
nature, that is, they are created and maintained
by contracts. Affective bonds, on the other hand,
such as mutual concern or love, are either adven-
titious, irrational, or morally irrelevant. The nuts
and bolts of the system are the contracts; the rest
is window-dressing or sentimentality. Only con-
tractual bonds are strictly morally binding. There
is no way in which affective bonds that are non-
contractual can be morally required or, indeed,
can enter into the moral picture at all.

Finally, it should be acknowledged that,
when all is said and done, the paradigm of con-
tractual agreements is the kind of commercial
contract that incorporates financial arrange-
ments of some sort or other or exchanges of goods
and services. In spirit, if not in intent and effect,
contractualism represents commercialism. As
such, it turns doctors into merchants and entre-
preneurs and patients into consumers and pur-
chasers. (The comparison with bakers is classic.)
In the end, perhaps, the commercial implications
of medical contractualism make it a theory for the
benefit of the rich that leaves the poor out in the
cold.

The Bracketing of Motivation. As I have al-
ready pointed out, according to the contract
model, motivation is irrelevant, or even excluded
as unnecessary or befuddling. To begin with, in
the original bargaining position, contractarians
like Rawls and Gauthier stipulate the condition
of "mutual unconcern" among the bargainers. Af-
ter that, the only moral requirement is that con-
tractees perform their contracted obligations, and
why they do so, their motives, are morally irrel-
evant. For contractualism, the motivation of the
contractees is relevant only because it tends to
make predictions about their performance more
reliable.

By all accounts, then, contractualism is a
hardheaded theory. It makes the motives of doc-
tors morally irrelevant as long as they keep their
side of the contract — for whatever reason. In
other words, the moral basis of a doctor's duty to
care for patients is not that patients need the care
but that the doctor is bound by a contract.

It follows that, apart from being stipulated in
some indirect way in the contract, any sort of ap-
peal on moral grounds to the needs or suffering of

Vol. 60 No. 1

CRITIQUE OF THE CONTRACT MODE^-LADD

a patient or to the requirement of sympathy, com-
passion, care, or concern on the part of the doctor
is not part of the doctor-patient relationship. If
doctors show such concerns, their doing so is su-
pererogatory and they are like heroes or saints
acting "beyond the call of duty." In this regard,
the doctor-patient relationship is comparable to
the buyer-seller relationship in a store: the buy-
er's motives are irrelevant as long as he pays, and
the seller's motives are irrelevant as long as he
delivers the goods. Friendly smiles and concerns
for the other are simply frosting on the cake or
good public relations. Giving something away for
free makes no sense on this model.

It should be clear by now that contractual-
ism, when applied to the conduct of doctors and
patient, is committed to a minimalist ethics, an
ethics that consists solely of strict obligations,
that is, fulfillment of rights, and acts of superer-
ogation, as for example acts of kindness. By the
same token, common garden-variety virtue is not
a moral requirement, for on a minimalist view
like contractarianism, a good doctor is simply an
honest one, who keeps his promises. There is no
obligation to provide "extras," like being a "nice"
or "friendly" doctor, much less to feel sorrow or
shed tears when a patient dies — unless, of course,
it is good for business.

The moral weaknesses of the contractual
model should now be amply evident. They may be
summed up by saying that it dehumanizes the
doctor-patient relationship and neglects the im-
portant and valuable aspects of that relationship
that revolve around caring for patients and heal-
ing them, activities that may be regarded as ends
in themselves rather than simply as means to ful-
filling a contract and making money.

We need another kind of theory.

Another Approach:
The Good Doctor and the
Ethics of Responsibility

In this final section, I want to move to an-
other level and try to sketch a different and I hope
more adequate view of the doctor-patient rela-
tionship. The new approach might be called a
"virtue ethics," although the conception of virtue
that I have in mind is considerably different from
traditional theories of virtue, like Aristotle's,
which identify it as a quality of persons (charac-
ter) rather than of conduct and relationships. The
kind of virtue ethics that I find most satisfactory
for analyzing the doctor-patient relationship fo-
cusses on the concept of a good doctor as a prac-
titioner, which in turn hinges on the quality of

the mutual relationship between doctors and pa-
tients and the mutual responsibilities that the re-
lationship implies.

The version of virtue ethics that I propose
here rejects the view that all duties are contrac-
tual or rights-based (that is, strict duties), and
holds instead that there are also duties of the
kind that Kant calls "duties of virtue," duties that
are defined by motivation, such as care, concern,
and compassion. These duties relate to other-re-
garding actions directed at ends such as the well-
being of others and the mitigation of suffering.
Health, of course, must be included here. Duties
of this kind arise in response to specific needs of
particular persons in specific interpersonal con-
texts like the doctor-patient relationship. Accord-
ingly, there are specific duties of virtue associated
with being a good doctor, that is, duties of care,
compassion, and concern in response to the needs
and sufferings of patients. The kind of duties in-
volved here belong under an ethics of responsibil-
ity. (Responsibility, as used here, is a virtue, just
as its opposite, irresponsibility, is a vice.)

To be a good doctor, therefore, means to take
responsibility for one's patient. Doing so is a re-
quirement of virtue as applied to the conduct of
professionals and others who have a special rela-
tionship to them, such as patients. As such, the
virtue ethics in question, unlike the Aristotelian
version, is situational and interpersonal. It might
be called modal.

The kind of virtue ethics proposed here is
egalitarian rather than elitist. For among other
things, a moral (or virtuous) doctor-patient rela-
tionship requires that doctors and nurses treat
patients as persons — ends-in-themselves, as Kant
would say. A patient is not just a body or a case,
or even a slave or an animal, to be fed, treated,
and ordered around. The relationship between
doctor and patient is a social relationship be-
tween persons as equals, which as such requires
mutual consideration, respect, trust, truthful-
ness, and, where feasible, shared decision-mak-
ing.

Thus it can be seen that, like contractualism,
a virtue ethics or, more particularly, an ethics of
responsibility preserves the principle of equality
and antiauthoritarianism. But unlike contractu-
alism, it centers more directly on the needs of the
patient and the care and concern of the doctor and
on the relationship that grows out of them. It
builds on the principle of humanity.

For all these reasons, as well as for many
others, I propose that we try to develop an
ethics of the doctor-patient relationship along the
lines of an ethics of responsibility, which empha-

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

sizes the tenor (mode) of the relationship, rather
than settling for a contractualist ethics simply
because it is assumed to be the only option left
after paternalism has been discarded.

References

1. Ladd J. Legalism and medical ethics. In: Davis JW,
Hoffmeister B, Shorten S, eds. Biomedical ethics. Hu-
mana Press, 1978. Also in J Med Philosophy 1979; 4(1).

2. Ladd J. The internal morality of medicine: an essential

dimension of the patient-physician relationship. In:
Shelp EE, ed. The clinical encounter. Dordrecht: Rei-
del, 1983.

3. Ladd J. The good doctor and the medical care of children.

In; Kopelman L, Moskop JC, eds. Children and health
care. Dordrecht: Kluwer Academic Publishers, 1989.

4. Vallentyne P, ed. Contractarianism and rational choice:

essays on David Gauthier's morals by agreement. Cam-
bridge: Cambridge University Press, 1991.

The Physician's Virtues and
Legitimate Self-Interest in the
Patient-Physician Contract

Laurence B. McCullough, Ph.D.

The bioethics literature abounds with discus-
sions of the patient-physician contract. Almost
exlusively, these discussions concern the physi-
cian's ethical obligations to his or her patient. In-
creasingly, ethical obligations of physicians to
third parties to the patient-physician relationship
must also be considered. The physician's ethical
obligations to his or her patients are typically un-
derstood in terms of the by-now familiar ethical
principles of beneficence, nonmaleficence, and re-
spect for autonomy (1). The physician's ethical ob-
ligations to third parties such as families of pa-
tients and the institutions that organize, deliver,
and pay for health care are typically understood
in terms of the principle of justice (1). This much
is familiar territory. Interestingly, it is a territory
still marked by a great deal of contention and
disagreement. This is as we should expect, given
the complexity of the patient-physician relation-
ship and its embeddedness in a complex web of
third parties. This much is clear: the physician is
the agent not only of patients but also of third
parties. The obligations that the physician has to
both must be carefully identified and negotiated
in a thoughtful manner.

It will come as no surprise that many physi-
cians conclude that this discussion is one-sided.
As one of my colleagues puts it to me, regularly
and with no little exasperation, "Physicians have

Adapted from the author's presentation at the conference
"Whose Contract Is It Anyway? Examining the Doctor-Patient
Relationship" at the Mount Sinai Medical Center on March 8,
1991. From the Center for Ethics, Medicine, and Public Issues,
Baylor College of Medicine, Houston, TX. Address reprint re-
quests to the author, Professor of Medicine and Medical Eth-
ics, Center for Ethics, Medicine, and Public Issues, Baylor
College of Medicine, One Baylor Plaza, Houston, TX 77030.

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

all of the obligations and patients all of the
rights!" Even a cursory reading of the bioethics
literature, not to mention even the slightest
brush with the law of malpractice, gives no other
impression.

It is long past time to redress this imbalance,
and I want to take some tentative steps in that
direction. I say "tentative" steps, because the ter-
ritory that I intend to chart is little explored and
little understood. The opportunities for error and
confusion, therefore, abound. As philosophers like
to say, there are many conceptual thickets in such
an uncharted territory. Let us see how far we can
explore without too substantial a loss of dermal
protection.

The Physician's Virtues:
Self-Effacement and
Self-Sacrifice

In my own view — and the view of many in the
field of bioethics — the patient-physician relation-
ship finds its ethical basis or foundations in the
character of the physician, in the physician's vir-
tues (2). The virtues are those habits or traits of
character that blunt one's self-interest and direct
one's concern to the interests of others. On this
basis, the physician's basic ethical obligation to
the patient can begin to be identified. That obli-
gation is to protect and promote the interests of
the patient and of third parties (3). The ethical
principles are necessary to fill in the detailed con-
tent of this general ethical obligation in concrete
clinical situations. This blunting of the physi-
cian's self-interest also inclines the physician to
fulfill his or her concrete ethical obligations to the
patient.

Two virtues seem fundamental. The first is
self-effacement, which blunts self-interest and fo-
il

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

cuses the concern of the physician on the interests
of patients and third parties. Self-effacement does
so by negating the adverse impact on the physi-
cian's attitudes and behavior of differences be-
tween the physician and the patient that should
not count in the care of the patient. Differences in
such matters as socioeconomic class, education,
race, gender, culture, religion, language, man-
ners, hygiene and — as the great physician-ethi-
cist of the Scottish Enlightenment, Dr. John
Gregory, called them — the "thousand peculiari-
ties to which the sick are subject" (4) are to be put
aside, erased from the physician's scope of con-
cern. The physician's concern is thus directed
away from mere self-interest — this patient is
drunk or smells or does not seem to understand
what one has to say — toward the interests of the
patient.

Taking care of patients is demanding and re-
warding work. Taking care of patients can also be
exhausting, and sometimes even dangerous. De-
manding work and personal risk call for sacrifice
on the part of the physician — sacrifice of time,
energy, obligations to others, and even health and
life. The latter sacrifices, by the way, were not
introduced by HIV or HBV — as any emergency
department or military physician over the years
could well remind us all. Self-sacrifice requires
physicians to accept risks to themselves and to
display a disciplined calm in that acceptance (2).
In this way mere self-interest is blunted in favor
of protecting and promoting the interests of pa-
tients and third parties.

Now, self-effacement and self-sacrifice can be
tyrannical virtues — and thus vices. That is, vir-
tues can be transformed into destructive traits of
character, if they are taken to be absolute, when
they are taken to admit of no limits in the moral
lives of physicians. Some have asserted such a
view. Pellegrino, in speaking to the obligations of
physicians to take personal risk in the care of
HIV-infected patients, writes: "To refuse to care
for AIDS patients, even if the danger were much
greater than it is, is to abnegate what is essential
to being a physician" (5). This sort of claim be-
trays a profound misunderstanding of self-sacri-
fice and of the virtues generally.

The virtues blunt self-interest but in doing so
they cannot eliminate all self-interest. That is,
not all forms of self-interest that might be op-
posed to the interests of patients and third parties
amount to mere self-interest or, more bluntly,
selfishness. The mistakes that Pellegrino makes
are two. First, he fails to recognize the category of
legitimate self-interest on the part of the physi-

cian. Second, he fails to recognize the compatibil-
ity of legitimate self-interest with the virtues of
self-effacement and self-sacrifice.

The Role of the Legitimate
Self-interest of the Physician

The physician's legitimate self-interest, I
submit, must be taken into account in the ethics
of the patient-physician contract. Physicians' le-
gitimate self-interest may even serve as the basis
of ethical obligations on the part of patients and
third parties to physicians, as I hope to show. Be-
fore exploring why this may be the case, however,
we first need to get clear about the concept of the
legitimate self-interest of the physician.

In the bioethics literature this concept has
not been accorded the careful consideration it de-
serves. And so we enter largely uncharted philo-
sophical and clinical territory. Legitimate self-in-
terest can usefully be understood to comprise a
continuum of ethically significant features of the
physician's moral life, considered as a whole.

Mundane Requisites for Good Patient Care.
First, physicians can properly claim a legitimate
self-interest in the requisites for good patient care
in their own day-to-day lives. I have in mind here
such apparently mundane, but in truth ethically
significant, matters as the time to study, reflect,
and learn and the time to rest and to maintain an
alert mind, to name only two that come quickly to
mind. That is, some forms of self-interest should
count as forms of legitimate self-interest when
those forms of self-interest can be tied to the nec-
essary conditions for providing good patient care,
day in and day out, over a lifetime.

Obligations to Persons Outside the Medical
Network. Second, physicians are bound by ethical
obligations to people other than patients or third
parties, for example spouses, lovers, friends, chil-
dren, and neighbors. To be sure, the physician
may have freely engaged in the relationships
within which such ethical obligations come to life.
However, the physician is not morally free to ab-
rogate those obligations without the permission
of those affected. Thus, the fulfillment of such ob-
ligations is a central feature of the physician's
moral identity as a whole person. Such ethical
obligations therefore properly constitute, in part,
the physician's legitimate self-interests.

Meaningful Nonmedical Activities. Finally,
legitimate self-interest also should include those
activities that the physician finds to be meaning-
ful outside being a physician and therefore out-
side the demanding network of obligations to pa-

Vol. 60 No. 1

PHYSICIAN VIRTUE AND SELF-INTEREST— McCULLOUGH

tients and third parties. These are the activities
xhat engage the physician as a "private" person
and that also provide him or her deep fulfillment
and satisfaction: fly-fishing, attending the opera.
They need not, but often in fact do, support and
sustain the two types of legitimate self-interest
that I have just described.

"Mere" Self-interest and the Virtue of Self-
lEffacement. Mere self-interest — "selfishness"
seems too harsh a term — comprises those forms of
self-interest that cannot be shown reliably to
count as one of three forms of legitimate self-in-
terest I have just described.

Self-effacement is directed as a virtue to
blunting forms of self-interest that are in almost
all cases, I think, forms of mere self-interest.
These mere self-interests — for example, petty
frustration with the so-called noncompliant pa-
tient or anger at bureaucratic hurdles which
third parties seem devoted to erecting in the phy-
sician's path — can be readily reined in by the vir-
tue of self-effacement. Thus, for the most part,
self-effacement is a virtue the moral demands of
which should be regarded in clinical practice as
routine and nonburdensome.
I Anger as Legitimate Self-interest. There
I may be some exceptions. Consider the following.
Earlier this week a colleague of mine, a pediatric
critical care physician, told me of how he had fi-
nally had it with a grandmother who was inter-
fering with her daughter's decisions about the pa-
tient, the woman's grandchild. This woman was
also interfering, sometimes quite obtrusively,
with the care of the child by nurses, physicians,
and technicians. She was a general nuisance most
of the time, something worse sometimes. Over
many days, he said, he asked this woman to ap-
preciate the difficulties she was causing, he tried
to understand her needs and respond to them, and
he asked her please not to do what she was doing.
Finally, he said, that morning, after a particu-
larly unpleasant event, he had told her, "Back
off!"

This, of course, was an expression of anger.
As such, it was clearly meant to focus this wom-
an's mind and change her behavior. One might
ask, are there justified expressions of anger on
the part of a physician? After all, is anger at the
surrogates of pediatric patients — or at patients
themselves — not a form of mere self-interest that
self-effacement is supposed to blunt? I am not so
sure, especially when such expressions of anger
are undertaken as last-ditch efforts to get some-
one's attention and change behavior that is at
risk for adversely affecting the patient. Some-

times, expressions of anger can succeed in accom-
plishing these goals when nothing else has.

If this line of reasoning makes sense, then
the following can be said with some confidence:
patients or patients' surrogates have no right to
take umbrage at legHimate expressions of anger
or other forms of legitimate self-interest rooted in
the well-being of patients on the part of physi-
cians. Indeed, patients and their surrogates are
ethically obligated to attend to these forms of le-
gitimate self-interest and act on them, one might
say. After all, it seems, patients and their surro-
gates owe their physicians a certain level of seri-
ous attention and consequent intellectual and be-
havioral discipline. This is especially true when
the stakes (of allowing undisciplined behaviors to
go unchecked) are high for patient care. It would
be worth inquiring at more length into other ex-
amples of legitimate self-interest that the virtue
of self-effacement should not blunt. But this is a
large topic and beyond the modest scope of this
presentation.

Negotiating Conflicts between Self-Sacrifice
and Legitimate Self-interest The virtue of self-
sacrifice also affects both mere and legitimate
self-interest. If the care of a particular patient
requires only a little extra time or if the hospital
requires paperwork to be properly completed so
that it can routinely collect payments and secure
the financial base of its mission, expecting the
physician to be self-sacrificing seems almost cer-
tainly to be reasonable. When self-sacrifice in-
volves risk to the physician's health or life, self-
sacrifice aims at something more, the regulation
of the physician's legitimate self-interest, because
risks to health and life involve all three forms of
legitimate self-interest. A question that must be
addressed in the ethics of the patient-physician
contract is. Must legitimate self-interest always
be sacrificed?

One thing seems clear from the outset: the
virtue of self-sacrifice cannot be understood to be
absolutely controlling of the physician's thought
and behavior, as Pellegrino would seem to have
it. This is because legitimate self-interest should
always be taken into account by the physician in
the process of determining the proper moral de-
mands of the virtue of self-sacrifice.

There is no easy algorithm for negotiating
conflicts between self-sacrifice and legitimate
self-interest. In part, this is because there is no
uniform set of legitimate self-interests that char-
acterizes each and every physician. Some are
married; some not. Some have or wish to have
children; some not. Some have weighty obliga-

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

tions to family, and may, for example, be caring
for a frail parent at home; some have routine ob-
ligations to family. Some are devoted full-time to
medicine; some cannot flourish absent an occa-
sional foray in a trout stream or an evening of
Puccini. As a consequence, the process of negoti-
ating conflicts between self-sacrifice and legiti-
mate self-interest will allow of some variability.

Some rules of thumb, though, seem reason-
able. First, every effort should be made to reduce
unnecessary impositions on legitimate self-inter-
est in day-to-day clinical practice. I have in mind
here such matters as making universal infection
control a reality, establishing rational and fair
staffing and call schedules. Second, the physician
needs to distinguish carefully between his or her
mere and legitimate self-interests. When in doubt
the physician should favor the conclusion that the
self-interest in question should be treated as mere
self-interest. Third, one's judgment of the resolu-
tion of conflicts between self-sacrifice and legiti-
mate self-interest should be tested for its intellec-
tual rigor and reliability against the considered
judgment of colleagues.

Fourth, institutional practices and policies
need to be carefully reviewed, to determine
whether they impose a tyrannical concept of self-
interest on the one hand or encourage mere self-
interest, for example in remuneration for its own
sake, to rule the roost. Both, I believe, are a
threat to the moral life of the physician. Institu-
tional third parties would therefore seem to have
an ethical obligation to protect and promote both
the virtues and the legitimate self-interests that
are essential to the moral lives of physicians. In
other words, institutional third parties have an
obligation to respect the autonomy of the physi-
cian when that autonomy is exercised on behalf of
legitimate self-interest. In my own view, institu-
tional third parties for the most part have yet to
acknowledge this as among their ethical obliga-
tions. We are all of us, patients and physicians
alike, at moral peril as the result.

Finally, a physician who reliably reaches the
conclusion that a conflict between self-sacrifice
and legitimate self-interest should be managed in
favor of legitimate self-interest should explain
the conclusion to the patient. After all, patients,
too, have a vital stake in sustaining the moral
lives of their physicians. Too, one might argue
that respect on the part of the patient for the au-
tonomy of the physician should include respecting
the physician's legitimate self-interests. This
would be a direct parallel to the physician's au-
tonomy-based obligation to respect the patient's
legitimate interests.

Summary

I will be the first to admit that we are now
well into uncharted territory of the patient-phy-
sician contract. I also detect missing stretches of
my dermal layer and you may spy some that I
have yet to notice. In any case, I put to your se-
rious consideration the proposal that part of the
patient-physician contract must include respect
for the legitimate interests of the physician by
patients and third parties.

The virtues of self-effacement and self-sacri-
fice and the concept of legitimate self-interest
help us to understand in concrete, clinically ap-
plicable terms what such respect means in prac-
tice. That respect will, I think, be expressed with
some variability, because there is no simple algo-
rithm for negotiating conflicts between legiti-
mate self-interest and the virtues of self-efface-
ment and self-sacrifice.

One important consequence of this moral
variability is that the patient-physician contract
and the virtues that sustain it will not yield to a
single, finally authoritative account of how such
conflicts should be negotiated. Instead, as we turn
more attention to these matters, we will, I be-
lieve, discover that there is a range or continuum
of ways in which the management of such ethical
conflict can reliably be understood in the patient-
physician contract. Rather than a single account
of the ethical dimensions of the patient-physician
contract, we should expect to develop a range of
reliable accounts. A kind of rich and engaging
moral pluralism should thus govern our under-
standing of the ethical dimensions of the patient-
physician contract. The interesting and impor-
tant issues will concern the boundaries of that
moral pluralism: which accounts of the contract
should count as ethically reliable and which
should not? I hope to have made a small start in
the direction of the larger and certainly more
daunting task of addressing this important ques-
tion.

References

1. Beauchamp TL, Childress JF. Principles of biomedical

ethics, 3rd ed. New York: Oxford University Press,
1989.

2. McCuUough LB. Ethics in dental medicine. J Dent Ed

1985; 49:219-224.

3. Beauchamp TL, McCuUough LB. Medical ethics: the

moral responsibilities of physicians. Englewood Cliffs,
NJ: Prentice-Hall, 1984.

4. Gregory J. Lectures on the duties and qualifications of a

physician. London: W. Strahan, 1772.

5. Pellegrino ED. Altruism, self-interests, and medical eth-

ics. JAMA 1987; 228:1939-1940.

Consumerism Rampant:

A Critique of the View of Medicine as a Commercial Enterprise

Daniel A. Moros, M.D.

Fhe title of this conference, "Whose Contract Is
'[t Anyway?" immediately reminds us that many
parties have an interest in the conduct of the re-
lationship between doctor and patient. In addi-
tion to the two principles, others include the fam-
ily, the state, insurers, medical institutions such
as hospitals and health maintenance organiza-
tions, and at times, the employer. Clearly the
more parties granted standing, the greater the
potential for conflict. Although we have always
known that in a fee-for-service system there is a
potential conflict between physician and patient,
in recent years our awareness of conflict within
the medical setting has been heightened by such
factors as the efforts of the state and third-party
payers to limit medical costs, and differing views
about the proper application of new technologies.

Since interests may conflict, we are chal-
lenged to characterize the doctor-patient relation-
ship and explicitly identify the legitimate expec-
tations and duties of the parties. It has certainly
proven easier to describe conflicts than to state in
positive terms what the doctor-patient relation-
ship ought to be. At first glance the title of our
symposium seems to presuppose that the lan-
guage of contracts will accommodate the reality
of this relationship. However, although the lan-
guage of contracts is remarkably flexible and can
be applied to a broad range of social interactions,
a question remains as to its adequacy for captur-
ing many essential features of the physician-pa-
tient relationship. In exploring this question we

Adapted from the author's presentation at the conference
"Whose Contract Is It Anyway? Examining the Doctor-Patient
Relationship" at the Mount Sinai Medical Center on March 8,
1991. From the Department of Neurology, Mount Sinai School
of Medicine. Address reprint requests to the author, Associate
Professor of Neurology, Mount Sinai School of Medicine, Box
1135, One Gustave L. Levy Place, New York, NY 10029.

must avoid unwittingly simplifying our view of
the physician-patient relationship in order to ac-
commodate our notion of a contract. Unfortu-
nately, this is a common error, and an overly sim-
ple view of the medical enterprise and the doctor-
patient relationship is presupposed by much
current bioethical discussion as well as health-
related legislation and case law.

Five Oversimplifications of the
Medical Enterprise

I start, therefore, by outlining five widely
held, interrelated perspectives about medical care
that I believe lead to a pervasive misunderstand-
ing of the medical enterprise. In critiquing these
five "simplifications," I must also convince you
that they are, indeed, widely held, and that I am
not attacking a straw man.

The Commercial Metaphor. The first simpli-
fication, which may be labelled the "commercial
metaphor," is the widely held view of medicine as
mere commerce and of patients as health-care
consumers. I suspect that within our society the
commercial metaphor is so powerful and en-
trenched, in the form of legislation, regulations of
federal and state authorities, case law, and insti-
tutional policies, that remarkably few clear and
explicit alternative images are available. Indeed,
I would suggest that as uncomfortable as many
physicians are with this commercial image, it has
nonetheless been implicitly embraced by the pro-
fession itself.

In support of this contention, consider the fol-
lowing passage from a U.S. Appeals Court deci-
sion of Feb. 1990 (1) upholding an earlier district
court's judgment in which the AMA was "found
guilty of a conspiracy to destroy the competitive
profession of Chiropractic":

Relief here is provided not only to the plaintiff chiroprac-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

tors, but also in a sense to all consumers of health care
services. Ensuring that medical physicians and hospitals
are free to professionally associate with chiropractors (e.g.,
by the publication and mailing of the order to AMA mem-
bers), likely will eliminate such anticompetitive effects of
the boycott as interfering with the consumer's free choice of
choosing a Product (health care provider) of their liking
(italics added).

Notice that there is no discussion here of pa-
tients and their needs or doctors and their du-
ties— only of consumers and their likings. The
commercial metaphor is not simply a matter of
the doctor as tradesman, or, to adopt the language
of the court, the doctor as "a product" — there is
also the image of the patient as a consumer of
medical products and services. Similar-sounding
passages can be found in Federal Trade Commis-
sion rulings eliminating restrictions on physician
advertising.

Linked with this commercial view of the phy-
sician-patient relationship is a rather simple no-
tion of the medical enterprise which has been
dubbed the "autonomy paradigm" by Jennings,
Callahan, and Caplan in their 1989 paper, "Eth-
ical Challenges of Chronic Illness" (2). In this pa-
per the authors critique "three interrelated no-
tions [or simplifications] that form the conceptual
infrastructure of most contemporary bioethics":

• A medical model of disease which views illness
"as a threat that suddenly intrudes upon a pre-
existing condition of health and wellbeing" (p.
9). "Here illness is seen as an alien threat to the
self, and the goal [of medicine] is to defend and
restore the self by curing or compensating for
the illness" (p. 8).

• An individualistic view of the person in which
"autonomy means freedom from external limits
or constraints. In this perspective, the autono-
mous self ... is independent of and prior to its
social milieu and its bodily condition" (p. 8).

• A contractual model of the physician-patient
relationship. According to this view "the pa-
tient is a rational, self-interested subject who,
threatened by illness, voluntarily enters into a
contractual agreement with a physician (or
other health care provider) and temporally sub-
mits himself or herself to medical authority in
order to combat the illness" (p. 8).

The authors maintain that these perspectives
have "become institutionalized in health care de-
livery through the influence of court rulings, gov-
ernment regulations, and the organizational pol-
icies of hospitals and other health care facilities."
Are they correct? Do these three perspectives in-
deed capture much of our society's view of the
medical enterprise?

The Medical Model of Disease. Consider the
medical model of disease. Something goes wrong,
you fix it, and that represents a service for which
third-party payers will reimburse. Someone has a
symptom, you examine the person and make a
diagnosis, and again the insurance company will
pay. You do not make the diagnosis but order an-
other test; they'll pay. Order five tests when one
would have done if only you knew how to take a
better history from the patient (or were willing to
devote the necessary time to obtaining a history),
and the insurance company will pay. Make the
diagnosis quickly and expeditiously, and they
will pay less. Write on the insurance form "nc
evidence of neurologic disease, would not do any
further testing at this time," or write "well-baby
examination" or "depression with somatization'
(assuming you are not a psychiatrist) and you will
often go begging for your money. Talk to a fright-
ened patient for 20 minutes at 10 pm or spend 3C
minutes in discussion with a patient's sibling whc
calls from San Francisco — forget about billing
This is the medical model of disease. You may be
judged for your competence and compassion, but
you get paid only for curing or active attempts to
do so.

A Highly Individualistic Notion of the Self.

How widely held is the highly individualistic no-
tion of the self? We increasingly see within oui
society an emphasis on conflict of interest be-
tween individuals, as opposed to bonds forged bj
shared experiences and common goals, hopes, and
values. We live with increasingly complex regu-
lations concerning child abuse, spouse abuse, el-
der abuse, and so on, as if interests can be defined
in isolation. I am not disputing the value of such
regulation. However, I am suggesting that our
relative ease in talking about rights, which em-
phasize the discreteness of individuals, and our
difficulty in talking of duties, which inevitably
emphasize the interconnections of individuals, is
part of this highly individualistic notion of the
self.

Viewing the Physician-Patient Relationship
as a Commercial Contract. As for the contractual
nature of the physician-patient relationship, I
think both the title of our conference and the lan-
guage of the Federal Appeals Court quoted above
testify to the pervasive influence of this simplifi-
cation.

Identifying a Statement of Preference with
Autonomy. The fifth simplification, which I dis-
cuss more fully below, is the tendency to discon-
nect the notion of autonomy from the notion ol
rationality and to focus instead on the mere state-
ment of a preference — that is, the identification of

Vol. 60 No. 1

CONSUMERISM RAMPANT— MOROS

a statement of a preference with an autonomous
decision.

Critique of
These Oversimplifications

Simplified descriptions are not necessarily
bad. Ultimately they must be judged on the basis
of how usefully they direct our thinking and
how seriously they mislead us. Although I hope to
illustrate some of the deficiencies of these simpli-
fications, I want first to emphasize that these are
not five random, disconnected perspectives, but
rather that they hang together and support each
other and are ultimately components of a gravely
deficient vision of medicine. The commercial met-
aphor requires a simple notion of medical services
(to wit, the medical model of disease), a simple
notion of the contracting parties (conveniently
provided by the highly individualistic notion of
the person), a limited notion of physician duties
(provided by focusing on expressed preferences
rather than needs or "what is right"). With this
vision in place, the physician-patient relationship
indeed appears to be a straightforward commer-
cial contract.

Shortcomings of the Medical Model of Dis-
ease. To demonstrate the inadequacies of this vi-
sion, let us begin with a critique of the medical
model of disease. For heuristic purposes I want to
broadly divide medical intervention into four cat-
egories: acute care and chronic care, each of
which is either curative or noncurative. The med-
ical model of disease applies, at most, to circum-
stances of acute care associated with relatively
rapid cure and misdirects our thinking about
most of the rest of medicine.

The doctor provides acute care to many indi-
viduals— young, presumably "normal" and
healthy; elderly, independent but fragile; persons
with ongoing, identifiable but at the moment qui-
escent illness; other persons struggling with ac-
tive disease and disability. In this latter group,
new difficulties may be part of the ongoing dis-
ease process (as when a person with pulmonary
disease develops an acute pneumonia) or some
unrelated problem. When a surgeon treats a
trauma or burn patient, a classic example of acute
care, and amputates a limb or devotes years to
reconstructive surgery to create a functioning but
disabled patient, the doctor can be viewed as act-
ing to preserve life, but certainly not as providing
a "cure." When a person treated for carcinoma of
the bowel suffers a relapse after appearing to be
disease-free for many years, the acute care and
apparent cure of the original treatment do not

suddenly become redefined as chronic care. And if
the disease is truly cured, the yearly follow-up
examinations and psychological interventions
(perhaps no more than reassurance) are certainly
not acute care, but cannot be seen as chronic care
for an ongoing disease. Indeed, acute care and cu-
rative treatment may require years of attention
and monitoring before the results can be known.
Similarly, preventive care such as vaccinations,
or the treatment of asymptomatic hypertension,
or the surgical removal of a lesion that has the
potential to become malignant are difficult to la-
bel as either curative treatment or chronic care.
Although individuals with both acute and chronic
conditions may be cured, the medical world does
not neatly divide into acute and chronic illness, or
curative and noncurative interventions.

Thus the dominance of the image of the phy-
sician as healer, as a person who cures acute ill-
ness, as opposed to a person of scientific knowl-
edge and technical skills who functions as a
rational agent at a time of all illness, belongs to a
bygone era when knowledge and doctoring were
the province of a priesthood or of "medicine men"
(3). This is not to say that the doctor should not
usefully exploit his or her priestly, all-knowing
persona for a patient's benefit. But science and
rationality, that is, proper decision making in
the service of a patient, are the defining charac-
teristics of good medicine.

Shortcomings of the Notion of a Discrete
Self. I offer two criticisms of the "highly individ-
ualistic notion of the self . . . [that views] self
identity, autonomy and interests as conceptually
prior to and independent of the encounter with
illness" (2). First, people often do not seem to be
such discrete individuals, and second, people
change under the impact of their experience with
illness.

Some examples: A physician taking care of
an elderly couple, perhaps a husband disabled by
a stroke and a caring wife stretched to her limits,
rarely thinks of them as individuals, but as a cou-
ple, and it can be difficult to determine where the
needs of one end and the needs of the other begin.
When people (patients) consider their interests,
they often think in terms of other people, so that
some patients will not ask for help specifically
because they are afraid it will be given too readily
and at too great a sacrifice. Some elderly people
have saved their money for the next generation
and have no intention of squandering their assets
on medical care. This is a phenomenon that we
encounter less frequently in a setting where a sig-
nificant portion of the cost of medical care is paid
by third parties.

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

It is also clear that people are changed by
illness and that the self is an elusive concept. This
applies to more than just the patient with Alzhei-
mer's disease or head trauma who is no longer
quite the same, or the patient with cardiac dis-
ease or chronic renal disease who may become
much more self-preoccupied and at some emo-
tional level less connected with family. Consider
the following two cases. In each, a busy active
professional who has never before been ill has an
acute myocardial infarction, and is focused on re-
turning immediately to work, but for very differ-
ent reasons. For the first patient, the job requires
an early return and to miss work now will involve
the loss of great opportunity. For the second pa-
tient, to be able to return to work is a way of
denying that anything serious has happened. In
the first case it may behoove the physician to aid
the patient in achieving an early return to work,
even at some additional risk to health. In the sec-
ond case, as the illness is recognized for what it is
and the denial confronted, the urgency of the pa-
tient's demands subside. In the second case,
which person is making the contract — the person
of day 1 who has never before confronted illness
or the person of day 5 who is beginning to con-
front things? The changing nature of the self as
the patient experiences and confronts an illness
and the difficulty in identifying a single person as
the physician's sole concern (as in the case of the
elderly couple) complicates the idea of clearly de-
fined contracting parties.

Shortcomings in Equating a Stated Prefer-
ence with Autonomy. The inadequacy of using a
stated preference as a guide for the physician's
behavior is easily illustrated in another setting.
Imagine a 17-year-old child with acute lympho-
blastic leukemia (a fatal disease when un-
treated) whose intelligent and successful parents
have always been skeptical about the traditional
medical establishment. After watching their
child's discomfort following two cycles of chemo-
therapy (a treatment that offers a greater than
50% chance of cure) they opt for the use of La-
etrile (which offers no help in avoiding certain
death). The adolescent agrees with the parents'
choice.

But the goal of patient and parents is sur-
vival. Perhaps they even state that if the doctor
could be certain of a cure, they would bear with
the chemotherapy. This is a situation in which
intelligent individuals have an expressed prefer-
ence about means — the treatment — that is incom-
patible with their clearly espoused goal — cure,
survival. By and large in our society, the clearly
stated preference is credited as an autonomous

decision, and the contradiction is ignored. But
what ought the physician to do? Should the phy-
sician support the choice of means, or do every-
thing possible to reverse this decision?

In this example the parents and patient seem
to be basing a decision on an incorrect assessment
of the facts. They believe, or hope, that Laetrile
will work. The question of what to accept as fact
leads to considerations in the philosophy of sci-
ence that are germane to ultimately defining the
role of the physician, but are well beyond the
scope of this presentation.

However, we may rewrite this vignette so
that the determining factor for the parents is not
a belief about medication but rather a concern
about the child's pain. What if both parents and
child say, "Enough," but the physician knows
that the discomfort is bearable because he has
seen other children survive and ultimately do
well? What if the parents and child are Christian
Scientists and do not accept treatment?

The question before us is not to what extent
the physician is able to impose his judgment, but
rather the logically prior question. To what ex-
tent should the physician cooperate with the ex-
plicit directions of the patient (and family)? This
concern with goals and the moral demands that
develop from the physician's understanding of sci-
entific fact goes beyond the range of any simple
contract for goods and services.

Abandoning the
Consumerist Image

If we abandon the consumerist image of a
doctor as an independent contractor in possession
of a presumed cure which he sells for a fee to a
presumed autonomous and discriminating
health-care consumer, we can return to a more
demanding notion of a professional in possession
of some special knowledge which is to be applied
thoughtfully for a patient's benefit. When we em-
phasize knowledge and introduce the moral re-
quirement that it be thoughtfully applied, the
doctor becomes responsible for making a correct
technological (medical) decision consonant with a
patient's needs and desires. In this setting there
may be several medically reasonable choices, and
a patient's actual needs or desires may or may not
correspond to his or her expressed wishes at a
particular moment. Further, there may be a con-
flict between a patient's goals and her or his as-
sessment of the best strategy to achieve them.

The physician's responsibility at times de-
mands paternalistic action, including a critical
initial assessment of whether the patient's state-

Vol. 60 No. 1

CONSUMERISM RAMPANT— MOROS

ment represents an autonomous choice. A physi-
cian may be obHgated to interfere paternalisti-
cally when the patient is an adolescent, or a
young and still developing adult who may never
have had the occasion to have seen others manage
illness and who now has insufficient capacity or
time to consider what must be confronted. Also, in
the emergency setting, a severely injured or in-
tellectually impaired adult may require immedi-
ate attention and be given little or no role in de-
cision making. Although to some extent such
paternalism may be thought of as occurring at the
behest of the patient, it is clear that the duties of
the physician in the emergency room do not de-
rive from a simple, explicit contract made directly
with the patient at the time of illness.
1; Furthermore, even if a sophisticated contract
'were able to articulate much or all of the physi-
cian's duties and the patient's claim on the phy-
sician's time and best efforts, there would remain
an unavoidable tension in the idea of contracting
for a paternalistic relationship. Indeed, in light of
all the considerations above, a fully articulated
contract might not be consistent with the degree
of discretion required for the satisfaction of a phy-
sician's duties.

Knowledge, Thoughtfully Applied: The
Trusted Advisor. All this does not erase the pos-
sibility that individual relationships may be
markedly structured by a set of understandings
between physician and patient that seem like a
straightforward contract. But even such a rela-
tionship exists under the umbrella of a broader
set of demands on the physician.

Take as an example an elderly patient who
does not want to solicit his children's assistance
regardless of the personal consequences. Perhaps
this decision places an additional burden on the
physician. The physician might disagree with
this decision, argue against it, but still honor it.
The physician might agree, but argue the point in
order to establish that the decision represents the
patient's considered judgment. In the latter cir-
cumstances the physician's effort might be nei-

ther contractually nor paternalistically required.
It most closely resembles the behavior of a trusted
advisor acting in a relationship involving a con-
siderable degree of intimacy. But is such a rela-
tionship included in the contract between a phy-
sician and a patient? And why does a physician
try to act like a trusted advisor to a new patient?
Is it just good citizenship? Is it another form of the
obligation derived from the possession of special
knowledge? Is the physician indeed obligated to
fulfill the function of trusted advisor, or merely
required to deliver a properly performed medical
task (such as the reading of a chest x-ray)?

Ultimately, the physician's behavior will
vary depending on the wishes and understanding
of the patient. Furthermore, both the patient and
the doctor-patient relationship may change with
time, and the duties of the physician may alter
commensurately. Thus, even when a more ex-
plicit, contractlike arrangement is agreed to at
the beginning, if the need for critical decisions
arises at a time when the patient cannot partici-
pate in reassessing the original understanding,
the physician may need to unilaterally decide
whether the original understanding is still in ef-
fect. This is a circumstance that the physician
should strive to avoid. I am not here trying to
extend the range of unilateral decision-making
by the physician. Rather, my purpose is to illus-
trate the limitation of the contractual model of
the physician-patient relationship and the mis-
representation of medicine embedded in the com-
mercial metaphor.

References

1. Wilk et al. v. AMA et al., Nos 87-2672 and 87-2777,

United States Court of Appeals for the Seventh Circuit,
February 7, 1990.

2. Jennings B, Callahan D, Caplan A. Ethical challenges of

chronic illness. Hastings Center Report. Feb/March
1988, special supplement.

3. Moros DA, Rhodes R, Baumrin B, Strain JJ. Chronic ill-

ness and the physician-patient relationship: a response
to the Hastings Center's "Ethical challenges of chronic
illness." Philosophy 1991; 16:161-181.

The Necessity and the Limitations of the

Contract Model

W. D. White, Ph.D.

It seems clearly established at law that physi-
cian-patient relationships are regulated and ad-
judicated by contract law (1). When a doctor's
gatekeeper answers the telephone and makes an
appointment for a person to become a patient, in
that conversation a contract is joined. In a similar
way, when a hospital admits a patient, a contract
is made.

At law, contracts can be express (and writ-
ten); or they can be implied (and sometimes spo-
ken). There is no legal distinction between ex-
press and implied contracts, once established.
They have equally binding contractual force. For
example, in each of the cases above, physicians
cannot legally abandon a patient.

In a doctor-patient relationship the establish-
ing of a contract also imposes on physicians other
legally enforceable duties — duties not established
as contractual duties, but duties flowing from the
contract. The shorthand idiom for these legal re-
quirements is that physicians must "exercise rea-
sonable skill and care" in response to their pa-
tients. These duties are imposed by law, but they
are usually defined and given concreteness by the
professional ethos of medicine itself. The law un-
dergirds the duty (2); but medical practice gives it
content and specificity.

When a patient gets hurt — or better still,
when she feels angry, abused, not respected, or
emotionally or otherwise abandoned or harmed —

Adapted from the author's presentation at the conference
"Whose Contract Is It Anyway? Examining the Doctor-Patient
Relationship" at the Mount Sinai Medical Center on March 8,
1991. From St. Andrews Presbyterian College, Laurinburg,
NC. Address reprint requests to the author, Distinguished
Professor of the Humanities at St. Andrews Presbyterian Col-
lege, at 117 Standish Drive, Chapel Hill, NC 27514.

she can seek redress for alleged violations of the
doctor-patient agreement through "breach of con-
tract" litigation (with the pros and cons of those
procedures). Or if she is hurt by alleged failure to
exercise reasonable care, she can seek compensa-
tion through tort law (with its particular pros and
cons). Breach of contract is essentially the failure
to deliver a specified service or good. Tort is an-
other thing: a negligent injury, or an injury suf-
fered by the patient, caused by an action or a fail-
ure to act by the physician, in violation of the
legally defined "reasonable care" (1, pp. 340-371;
2, pp. 253-266; 3).

This arrangement in the law reflects the ju-
dicial recognition that contract law, and the
model or paradigm informing it, is not a fully ad-
equate model for addressing the complex of hu-
man problems exhibited in modern medicine — in
modern, technological medicine, which is at once
a tragic profession and a moral art (4, 5).

The law recognizes that medicine is not a
mere entrepreneurial enterprise — indeed, that it
is not in its essence or intentionality a business
transaction at all. Going to one's doctor is not
analogous to buying a used car. What is happen-
ing in the medical clinic is not an arms-length
transaction — or it should not be. "Caveat emp-
tor," "Let the buyer beware," the ruling motto of
the marketplace, has no place in the tragic pro-
fession of medicine, the practice of which is a
moral art. "The essential nature of the associa-
tion between practitioner and patient involves a
professional service rather than a sale" (2, pp.
261).

The law itself clearly recognizes all this,
which is why certain kinds of contracts that
might theoretically be entered into freely by mu-
tually consenting competent adults, without coer-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

Vol. 60 No. 1

CONTRACT MODEL: NEED AND LIMITS— WHITE

cion or deception or duress — where each party
voluntarily chooses, out of self-interest, to pay the
price and realize the benefits — are not legally
permitted in medicine (1, pp. 341-351). I have in
mind, for example, that you cannot contract for
less than "standard" medicine. You cannot shop
around for "bargain-basement" medicine that
fails to reach the accepted standard, however eco-
nomically feasible or desirable such an arrange-
ment might appear to be. Nor can you, through
contract, before the event, release hospitals and
physicians from liability for negligent injury aris-
ing from substandard care. Nor is it likely that
any physician could be held by contract with a
patient to achieve a specified therapeutic outcome
(2, pp. 253-259). These, I take it, are significant
signs from the law itself that contracts, while nec-
essary, are not fully adequate; and are some-
times, indeed, counterproductive as a model for
fostering humane and ethical medicine.

Why Contracts Are Needed

Contracts are necessary for a number of rea-
sons; and for these reasons contract law is appeal-
ing to many ethically sensitive persons (6). I
merely mention, or catalog, some of these reasons
in passing. First, the radical imbalance of power
between the sick person and the physician in the
medical clinic cries out for some redress, an end
which contract law envisions to some degree. The
"patient's rights" movement, in the wider context
of the various consumer and human rights move-
ments, lends impetus to the legal equalizing of
power. Contracts tend to specify rights and duties
and conditions which can be, if necessary, legally
enforced.

This seems appealing, particularly to a cul-
ture characterized by radical pluralism. We no
longer have (if we ever had) any widespread con-
sensual views of what human beings are, of what
the good life is, or of what our duties and privi-
leges toward each other are. Where there are no
assumed first principles with reference to such
basic matters, putting oneself in the hands of the
medical profession can be perceived as a danger-
ous enterprise.

And indeed it can be. These dangers are the
undertow of the great technological advances in
medicine in the last half century or so. Because of
the powerful, diversified invasive technologies
now available to physicians — therapies and pro-
cedures that can significantly help catastrophi-
cally ill persons, but which can often equally cat-
astrophically hurt them — many patients are not

content merely to say "Do something, Doctor," or
"Do what you think is best." Contract law seems
attractive because it suggests the necessity for
joint decision-making; in its nature it opposes
unilateral choosing. It invites physicians and pa-
tients alike to respect the personal and the pro-
fessional autonomy of the other, thereby encour-
aging discussions leading to "informed consent"
rather than "blind trust" from patients.

In these ways contract law tends to challenge
the traditional authoritarian models of the phy-
sician as parent (or as priest) simply doing "what
is best for the patient" — a tradition that no doubt
encourages patients to become or act like chil-
dren. Historically, medical paternalism at its best
has been a kind of idealized altruism; at its worst
it has shown high-handed hubris and arrogance.
No doubt this historical reality is a key element
in the widespread consumer demand that physi-
cians be legally accountable for what happens in
medicine.

Perhaps it should also be emphasized that in
the nature of our modern, often urban society —
and in the nature of tertiary-care hospitals and
other medical institutions serving wide sections
of the populace — medical relationships will be es-
sentially between strangers. Physicians taking
care of strangers is the norm rather than the ex-
ception in much of modern technological medi-
cine. This is another reason contract law is nec-
essary, and is often seen as appealing.

Limits of the Contract Model

For all these and a number of other reasons
that might be adduced, contracts are a necessary
(and on the surface, appealing) model for under-
standing legally what is happening in the medi-
cal clinic. But they are nevertheless inadequate;
they are limited in value. Like most of the legal
presences in the medical world, they promise
more than they can deliver. Although it is per-
haps understandable that a litigious society such
as ours turns to the law to try to bring some moral
order into the cultural chaos that medicine re-
flects, we nonetheless need to emphasize and to
reiterate that the law cannot finally help us much
with the wider and deeper ethical issues. Declar-
atory judgments in the law give the illusion of
making final, precise, and clear rulings. But the
fact is that this is an illusion. Law cannot elimi-
nate the ambiguity, contingency, and lack of clar-
ity, precision, certitude, and finality that exists in
medicine. The law is too blunt an instrument to
help us much in these matters.

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

In the same way that we need to deflate the
public expectation that medicine can become a
cure for death, we need also to challenge the pop-
ular notion that the active presence of the law as
a third party in the clinic will bring us humane
and ethical medicine. We must confess the ambi-
guity, celebrate the contingency, and criticize the
insistence on legal certitude, finality, precision,
and clarity in medicine. The public expects more
than the law can deliver. Upon careful reflection,
one might indeed conclude that the least law nec-
essary is the best law possible. But contract law
in medicine is necessary.

Yet insofar as medical ethics is concerned,
contract law is woefully inadequate in contribut-
ing to the creation and support of a humane and
ethically sensitive medicine (6, pp. 118—127).
Contract law is essentially limited in that it is
minimalist in its ethical demands. Self-interested
minimalism is the norm in contractual arrange-
ments. It is inadequate also in that the specifics of
a contract would always fail to recognize the full
scope of open-ended, unpredictable professional
ethical duties in a doctor-patient relationship.

The Gift of Presence. Furthermore, as
William F. May (6, p. 118) has persuasively ar-
gued, the model of contract "suppresses the ele-
ment of gift in human relationships." And "gift"
is an element which, despite its abuse in the his-
tory of medicine, is nevertheless central to taking
care of patients. While doctors have sometimes
engaged in the "conceit of philanthropy" (May's
phrase), seeing themselves only as givers, the el-
ement of giving and receiving is at the heart of
the phenomenon of sick persons and their physi-
cians making themselves present to each other in
their common mortality. Being personally
present — the "gift of presence" — is a crucial part
of humane medicine. And the contract paradigm
tends to obscure this dimension of gift which is at
the heart of humane relationships of every kind.
Hence, May concludes: "The kind of minimalism
that a purely contractualist understanding of the
professional relationship encourages produces a
professional too grudging, too calculating, too
lacking in spontaneity, too quickly exhausted to
go the second mile with patients along the road of
their distress" (6, p. 120).

Finally, contracts (though necessary) are
also, as a dominant paradigm, not only inade-
quate, but in some cases actively counterproduc-
tive of ethical sensitivity. For the contract tends
to move the ethical question in medical clinics
from "what should or should not be done" to
"what is proscribed by law; what is required by

law; what is permitted, but not prescribed, by
law." When physicians stop asking, "What is eth-
ically correct?" and ask instead, "What is the
law?" we have suffered a moral loss in the ethos of
medicine.

Contract as a model is also potentially coun-
terproductive in that it tends not only toward a
lower standard of care, but (paradoxically) it also
increases the temptation for doctors to practice
"defensive medicine." To protect their own inter-
ests, and out of anxiety before the law, doctors can
be tempted to do tests and follow procedures not
medically necessary or indicated. One might ar-
gue indeed that seeing medicine primarily under
a contract model has, indirectly, contributed to
and exacerbated the so-called malpractice crisis.

A More Compelling Paradigm:
Fiduciary Covenant

To summarize: contract law exists; it is nec-
essary and in many ways appealing. It is not ad-
equate, and it is potentially counterproductive as
a paradigm for understanding ethics in the med-
ical clinic. Can we do better? Is there another
more promising model we can invoke? I think we
can do better; and there is a more compelling par-
adigm.

I want to suggest that the controlling meta-
phor for envisioning medicine be not contract, but
rather covenant, fiduciary covenant. I believe this
incorporates the positive values of contract, but
goes beyond to correct some of its deficiencies.
The fiduciary covenant model describes more ac-
curately what is happening when physicians and
sick persons (and the worried well) make them-
selves personally present to each other in the di-
agnostic and therapeutic encounter. This model is
phenomenologically and descriptively more accu-
rate; and at the same time it holds up an ideal
against which current practice can be challenged
and judged. This model is historically rooted in
the religious traditions of Judaism and Christi-
anity; but I believe its descriptive and normative
power does not ultimately depend on theological
assumptions. The earliest significant use of this
model as a paradigm for medicine is by Paul Ram-
sey (8). William F. May further elaborates on the
covenant metaphor in a 1975 article (9), but his
most elaborate use of it is in his The Physician's
Covenant (6), which elaborates on the significance
of image and metaphor in biomedical ethics.

Veatch develops at length a social contract
theory that attempts to define a kind of secular

Vol. 60 No. 1

CONTRACT MODEL: NEED AND LIMITS— WHITE

paradigm incorporating some of the better fea-
tures of the reHgious model of covenant (7).

As the designation implies, the fiduciary cov-
enant model is rooted and grounded in trust. Not
the childlike trust which pleads with the kindly
parent figure, "Do something, Doctor!" Not the
blind trust which abdicates all adult responsibil-
ity and acquiesces, "Whatever you think is best.
Doctor." Not that kind of abject trust which seems
so often to have broken down when we hear com-
plaints that a patient is not "complying with doc-
tor's orders." No — not that kind of trust at all —
but a trust that is mutual, that respects the
person of the other, that recognizes the needs of
the patient for help, and the needs of the doctor to
have a patient, with all that that implies for each.

Such a fiduciary covenant, a covenant of
trust, is grounded in response to gifts received.
First, the physician's response to the gifts made to
her by society: the gifts of medical education, of
research facilities, of hospitals and clinics, of li-
censure and the authority to diagnose and pre-
scribe— in short, the gifts of a profession, apart
from which medicine could not exist. Covenant is
grounded in the physician's response to such gifts.
And it is also rooted in the patient's response and
need as the physician offers her gift of personal
presence, of expert knowledge, of clinical exper-
tise, and of the socially bestowed authority to pur-
sue the health of the patient.

These responses to gift involve the making
and keeping of promises, explicit and implied, by
both parties. Hence fidelity to the legitimate ex-
pectations of the relationship becomes the litmus
test of how ethical and humane the relationship
is. Such a model recognizes (and articulates) the
irreducible interdependence of human beings; it
makes short shrift of all forms of high-handed au-
thoritarianism, and of self-sufficient individual-
ism. It becomes a check on heavy medical pater-
nalism, however well-intentioned; and it is a
brake against ideological claims that the wishes
of the patient should have primacy or be absolute
in decision-making. The covenant model insists
that unilateral decisions are always ethically
faulty, whatever their substance and whatever
their consequences. It is a model that requires
physician and patient alike to bear the burden of
joint decision-making, allowing neither to aban-
don the other by withdrawing into silence. In all
these ways it affirms a high ethical ideal against
which concrete cases might be judged or evalu-
ated.

(If physicians had time and interest to read
only one book dealing with doctor-patient rela-

tionships, I would strongly recommend Jay Katz's
The Silent World of Doctor and Patient [101.)

Such a paradigm has profound implications
for how the personal presence of physicians and
patients to each other will be articulated. It is
essentially dialogical. It requires not only the "in-
formed consent" of the patient, as that new and
(to many physicians) novel idea has been devel-
oped in ethics and in law. But it also requires that
physicians articulate some of their own anxieties
and fears — including the medical uncertainty
they have to deal with every day, and their per-
sonal medical prejudices as well. Within the con-
fines of distributive justice (considering that a
physician's time is a "limited resource in high de-
mand"), and in the context of the individual pa-
tient's needs and desires, such an "uncovering" of
the "physician's burden" could be a salutary
thing. It could indeed open up clogged channels of
communication; and perhaps even more signifi-
cantly, it could help physicians themselves deal
with the ambiguities and uncertainties of their
daily routine. Traditional justifications for physi-
cians' always appearing "certain" and "in charge"
do not hold up persuasively under the scrutiny of
the fiduciary covenantal model (10, pp. 165-206).

Trust in such a covenantal relationship is
earned and reenforced by the making and keep-
ing of promises, all in response to gift and in the
context of need. It is articulated as mutual inter-
dependence by physicians and patients who are
personally present to each other in the common
pursuit of the intrinsic telos or goal of medicine,
which is the health of the patient. If the "well-
working of the organism as a whole" cannot be
achieved, the goal becomes the mutual interde-
pendence of physician and patient in caring for
the patient in the disintegrating of the organism.
More attention needs to be given to the appropri-
ate teloi of medicine — the ends intrinsic to medi-
cine, and integral to its practice (11).

This is the ideal which I believe we should
keep before us as we try to go beyond the mini-
malism of the contract paradigm. Fiduciary cov-
enant paradigmatically captures the complex
richness and unpredictability that lie at the heart
of humane doctor-patient relationships.

References

1. Wadlington W, Waltz JR, Dworkin RB. Law and medi-

cine: cases and materials. The Foundation Press, 1980;
102-170.

2. King JH, Jr. The law of medical malpractice. West Pub-

lishing, 1977; 39-54.

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

3. Quimby CW, Jr. Law for the medical practitioner. Aupha

Press, 1979; 55-82.

4. Hauerwas S. Truthfulness and tragedy. University of

Notre Dame Press, 1977; 184-202, 241-246.

5. Hauerwas S. Suffering presence: theological reflections on

medicine, the mentally handicapped, and the church.
University of Notre Dame Press, 1986; 23-83.

6. May WF. The physician's covenant: images of the healer

in medical ethics. The Westminster Press, 1983; 117-
118.

7. Veatch R. A theory of medical ethics. Basic Books, 1981.

8. Ramsey P. The patient as person. Yale University Press,

1970.

9. May WF. Code, covenant, contract or philanthropy. Hast-

ings Center Report 1975; (5):29-38.

10. Katz J. The silent world of doctor and patient. The Free

Press, 1984.

1 1 . Kass LR. Regarding the end of medicine and the pursuit ol

health. The Public Interest 1975; 40:27-29.

The Physician's Experience:

Cases and Doubts

Alisan B. Goldfarb, M.D., Thomas H. Kalb, M.D., Bennett P. Liefer, M.D., and Audrey Schwersenz, M.D.

Breast Surgery
Alisan B. Goldfarb

I am pleased to participate here so that I can
question in public all the things I normally ques-
tion in silence. I am going to discuss two patients.
Case 1 raises a broad issue that I deal with almost
every day during office hours. Case 2 raises many
different moral and ethical dilemmas for me as a
physician.

Case 1: When Nobody Knows the Right An-
swer. A 51-year-old woman had a mammogram
revealing abnormality, a few calcifications, a sub-
tle change from her previous film two years ago.
On biopsy, the change proved to be mostly fibro-
cystic breast tissue; however, there were several
microscopic cells of intraductal carcinoma,
j In breast surgery there are lots of choices,
depending on the diagnosis. Some of those choices
exist because the different medical treatments
are equal, and some of those choices exist because
we really do not know what is the right treat-

Adapted from the authors' presentations at the conference
"Whose Contract Is It Anyway? Examining the Doctor-Patient
Relationship" at the Mount Sinai Medical Center on March 8,
1991. ABG's presentation is from the Department of Surgery,
Mount Sinai Medical Center. Address reprint requests to the
author at 3 East 74th Street, New York, NY 10021.

THK's presentation is from the Division of Pulmonary
Medicine, Mount Sinai Medical Center. Address reprint re-
quests to the author at Box 1232, Mount Sinai Medical Center,
One Gustave L. Levy Place, New York, NY 10029.

BPL's presentation is from the Department of Geriatrics,
Mount Sinai Medical Center. Address reprint requests to the
author. Clinical Instructor in Geriatric Medicine, 130 Pros-
pect Street, Ridgewood, NJ 07450.

AS's presentation is from the Department of Pediatrics,
Mount Sinai Medical Center. Address reprint requests to the
author, Associate in Pediatrics, Mount Sinai Medical Center,
Box 1198, One Gustave L. Levy Place, New York, NY 10029.

ment, because we have no idea what the natural
history of the disease would be. The latter was
true of this case: these microscopic clusters of ma-
lignant cells indicate a new entity. Because the
mammogram machines have become so sophisti-
cated, microscopic abnormalities are found. Left
alone, they might never turn into a problem.
However, it is difficult for a physician to look at a
few malignant cells and believe that.

So the patient came to me for a third opinion.
The first opinion had been to do a mastectomy,
the second that she should do nothing but careful
follow-up. She came to my office not because she
wanted a tie-breaking vote, but because she
heard that I would take some time and try to ed-
ucate her about why she had gotten two such dis-
parate opinions.

What I was faced with, of course, and what I
am faced with often in this dilemma, is that no-
body knows the right answer. Most of our infor-
mation comes from extrapolation from diseases
that are similar, but not the same. Each of us has
our own emotional bias. At most meetings of
breast surgeons, we end up in hot debate. So, in
educating a patient, particularly an intelligent
and well-educated patient, do I have to be careful
not to lean toward my biases? If my bias is based
on what I consider to be a good medical education
and many years of experience, should I not be
biased?

The other question is, when I am educating
the patient, can I be sure that she is understand-
ing what I am telling her? If she chooses what I
would feel is the better course of action, and I am
not certain that she has understood, do I really
have to make her understand? If she is choosing
what I do not think is the better course and I do
not think she has understood, do I have to make
her understand?

If she turns around and says to me, "Dr. Gold-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

farb, what do you think is best? What would you
do?" I find the questions unanswerable because it
is not me, I do not know what I would do, and I
really do not know what is best. I know I have
feelings about what I think is best.

Such consultations are tremendously diffi-
cult. Generally, when you are giving the third
opinion, you cannot say "Well, you could talk to
other doctors and see what they say." I saw this
patient about two months ago, and she has not yet
made up her mind. She has not seen more physi-
cians, but she has consulted other ancillary spe-
cialties, including "alternate care" specialists, in
trying to make up her mind.

Case 2. When Complex Medical and Psycho-
social Developments Intersect. I first met this pa-
tient in 1986, when she was 27 years old, and
came to me for evaluation of a peculiar thicken-
ing in her left breast. She came to my office with
her mother who, in fact, had been operated on for
breast cancer in 1981. Although this young lady
was an adult, and married, the mother did most of
the talking at the initial office consultation.

Highly significant in the patient's medical
history was that, at age 11, she was treated for
Hodgkin's disease, originally diagnosed in a su-
perclavicular node. She underwent extensive
study, including staging laparotomy, and radia-
tion therapy to the test area as well as the ab-
dominal area. She was well for several years;
then a peculiar enlargement of her tonsils was
noted. The Hodgkin's disease had recurred in her
tonsil area and she was again treated with radi-
ation. After that, she was entirely well. Between
her adolescent years and the present, she did un-
dergo one lymph node biopsy, which was nega-
tive.

Either because of her previous illness, or its
treatment, or the emotional problems that go
along with them, the patient herself was quite
frail, both in physique and emotional presenta-
tion. The idea of biopsying the breast was some-
thing that she went along with and tolerated eas-
ily. Unfortunately, it did prove that she had a
small malignancy of the breast and although
other options seemed perhaps appropriate, the de-
cision was to do a mastectomy. She tolerated the
surgery well and the prognostic indicators were
favorable. She required no chemotherapy. She
and her mother decided that she would like to be
followed up by a medical oncologist, and they
chose someone who in personality fit their needs.
My feeling was that, since she did not require
chemotherapy but careful follow-up, the choice
was a good one. He was a rather grandfatherly,
paternalistic sort. She was carefully followed up.

Then, of course, the question of whether she
should become pregnant arose. The issue in my
discussing this with her went beyond the breast
cancer, because I had her well-meaning but over-
bearing mother saying to me, "She can barely
take care of herself. Tell her she can't tolerate it."
I had my own questions about whether a patient
with breast cancer should sustain the high levels
of hormone stimulation that occur in pregnancy. I
also had the real question of whether this patient,
who appeared so frail and snapped back physi-
cally from an operation so slowly, would deal well
with the demands of pregnancy and taking care of
a young child. The patient independently decided
not to become pregnant, or even to consider preg-
nancy.

The next thing I knew a letter arrived on my
desk from an adoption agency asking me whether
this patient was fit medically and emotionally to
adopt a child. My questions became complicated.
This was the young lady who (and I had breathed
a sigh of relief) had decided on her own not to
become pregnant. Could I write that emotionally
and physically she was a good candidate to adopt
a child? Was my responsibility to the patient, who
I knew would like to be a parent? To the adoption
agency? Or to the child ready for adoption? A
complicated problem. The paperwork, of course,
churned on and on.

During this patient's follow-up, a thickening
in the other breast was noted, which was biopsied
and proved to be benign. Unfortunately, our
breath of relief was short-lived because soon
thereafter a lymph node on the other side was
noted to have metastatic breast carcinoma. A sec-
ond primary lesion was found on the other side, so
this patient had to undergo a second mastectomy.
Unfortunately, this malignancy was much more
aggressive, and chemotherapy was required.

I was concerned that the original choice of
the medical oncologist, the grandfatherly type,
who was gentle and appropriate in dealing with
this patient and who, in fact, had been the one
who found the metastatic node in her axilla, was
not aggressive enough to give this young woman
the complicated treatment that she needed. The
radiation therapy had left her with some disabil-
ity that would make the chemotherapy much
more difficult. Luckily, he graciously phoned me
before deciding on what treatment to choose, and
together we decided for consultation with one of
the young, aggressive chemotherapists in the city
with whom he said he would be happy to work.
This would maintain his relationship with the pa-
tient and allow her the more aggressive chemo-
therapy.

Vol. 60 No. 1

THE PHYSICIAN'S EXPERIENCE— GOLDFARB ET AL

Then the mother came up with more advice.
"This is going to be very aggressive treatment,
don't you think she should move in with me while
she has it?" And the patient, trying to save her
marriage, which was in some difficulties, said, "I
live in New Jersey. Can't I be treated in New
Jersey?" My position then became not breast sur-
geon, but more advisor-marriage-counselor-psy-
chiatrist.

Another issue was raised in this patient's
case. The patient had always experienced some
heavy bleeding with her periods, and was known
to have a fibroid uterus. Once the difficult chemo-
therapy was begun, her gynecologist suggested,
"Since you will not be having children anyway,
and heavy bleeding might be a problem while
you're on chemo, let's do a hysterectomy before we
begin the chemo." The patient was quite tearful
as she reported this information to me. I had
great difficulty with this particular issue, because
I saw this lovely young lady, who had already lost
both breasts, facing the loss of the last of her fe-
male organs. Having seen many, many patients
through chemotherapy, I felt strongly that her
periods would probably end anyway after a cycle
or two of chemotherapy. I felt that this advice by
her gynecologist was unfeeling. I wrestled with
the question of my place in talking with the pa-
tient. Should I call the gynecologist? Should I call
the medical oncologist and have the medical on-
cologist call the gynecologist? Or should I speak
with the patient, or even her mother, about
whether they should insist on a dialogue between
gynecologist and oncologist?

Intensive Care Units
Thomas H. Kalb

A recent incident in the medical intensive
care unit in which a patient temporarily refused
medical treatment highlights for me some practi-
cal and frequently encountered issues regarding
the patient-physician contract and points up some
features of the ICU environment which bring
unique ambiguities and new twists to bear on this
contract.

A 39-year-old man who works as a hospital
administrator (at another institution) and carries
the underlying diagnosis of alcohol-related liver
failure with cirrhosis in 1990 had a lengthy stay
in The Mount Sinai Hospital for complications of
cirrhosis, including life-threatening bleeding gas-
tric and esophageal varices, for which he under-
went emergency surgery — a distal splenorenal
shunt.

His postsurgical course was complicated by
the development of progressively tense ascites, a

buildup of fluid in the abdominal cavity. To re-
lieve discomfort, a catheter was inserted to drain
this fluid. Of note, the ascites was milky, with a
relatively elevated triglyceride level, and this
analysis, coupled with the temporal relationship
to surgery, was consistent with the possibility
that abdominal lymphatics had been interrupted
at laparotomy, resulting in spillage of lymph fluid
into the abdomen — in other words, a surgical
complication. It was not clear from the chart rec-
ord whether this possibility was discussed with
the patient, although the patient's fiancee, who
described herself as having previously worked as
a nurse, assumed with evident displeasure that
this was the case. Moreover, the amount of fluid
which could be removed safely was limited by the
development of renal insufficiency, reflecting the
sensitivity of the kidney to shifts in body fluid
volume in the setting of cirrhosis.

After nearly three weeks of limited success in
dealing with these problems, the patient was dis-
charged on a regimen of fluid restriction and di-
uretics, with residual though stable ascites and
moderate renal insufficiency.

One week after discharge, he was readmitted
to the medical service for increasing abdominal
girth and lethargy, and on the second hospital
day was transferred to the medical intensive care
unit for recurrent gastrointestinal bleeding. In
consultation with the admitting gastroenterolo-
gist, the ICU team — consisting of attending phy-
sician, fellow, and medical resident — proceeded to
initiate several therapies to stabilize the patient.
These therapies included the reinsertion of an ab-
dominal catheter for drainage of ascites. Based on
his condition, it was agreed that a relatively
small volume of fluid would be drained, and that
thereafter, the catheter would be clamped.

The patient was lethargic but easily arous-
able and oriented, and he was accompanied by his
fiancee throughout the evening. The procedure
was described to both of them by the medical res-
ident, they agreed to it, and consent forms were
signed.

In the evening, the ICU is staffed by a med-
ical resident alone; a backup attending physician
and fellow are available on call. When the cover-
ing resident returned to the patient's bedside that
evening to clamp the abdominal catheter, she was
stopped by the patient's fiancee, who insisted that
the catheter not be clamped, saying (rather color-
fully) that she would prevent all intervention un-
til the attending gastroenterologist was con-
sulted, adding, "I don't know you from Adam."
The resident attempted to reexplain the indica-
tions, but both patient and fiancee steadfastly re-

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

fused. The resident contacted the ICU fellow and
the ICU attending, but the patient's fiancee re-
fused to come to the phone, stating that she sus-
pected a ruse to leave the patient unattended so
that the catheter could be clamped. Observing the
ruckus, the nursing staff independently notified
the administrative supervisor, who instructed the
resident by phone that the catheter not be
clamped against the family's wishes, and recom-
mended that the attending gastroenterologist be
notified. After two hours, the gastroenterologist
contacted the unit, and with a few reassuring
words to the fiancee, she finally agreed to contin-
ued medical care by the ICU staff. In the interim
standoff, nearly three liters of unintended drain-
age occurred. Fortunately the patient suffered no
immediate untoward effects attributable to this
incident, although his subsequent course was
rocky indeed, culminating one month later in suc-
cessful liver transplantation. Throughout his stay
in the ICU, EG's fiancee never flagged in her vig-
ilance, though as far as I am aware, no further
incidents, refusals of care, or other untoward
events between family and staff occurred.

When Physicians' Responsibilities Are Shared.
Several questions relevant to the patient-physi-
cian contract are, in my mind, raised by incidents
like this:

• Who is responsible for patient care in an ICU?

• Who do patients and their families think is re-
sponsible for their care when they have been
transferred to the ICU?

• What are the problems inherent in responsibil-
ity shared between ICU attending physicians
and staff, and private attending physicians?

Elderly Psychiatric Patients
Bennett P. Leifer

An 86-year-old white man was admitted to
the geropsychiatry service for the treatment of an
exacerbation of his chronic depression.

Significant in his medical history were non-
insulin-dependent diabetes, chronic renal insuffi-
ciency, Parkinson's disease, ischemic heart dis-
ease, and benign prostatic hyperplasia.
Significant in his psychiatric history were previ-
ous episodes of depression successfully treated
with electroconvulsive therapy.

The patient was a retired engineer, married
to a psychiatrist for about 60 years. He and his
wife lived on the Upper East Side of Manhattan
and always enjoyed the diverse cultural life of the
city. His wife had been diagnosed as having de-
mentia of the Alzheimer's type five years before.
She lived in their apartment with 24-hour super-
vision. They had no children and had outlived all

immediate family. They did, however, have a
close circle of friends, all in their late 70's. Before
the patient's wife developed dementia, she and he
had been active in their community.

The patient progressively deteriorated dur-
ing his stay on the psychiatric unit. Not only was
he unresponsive to electroconvulsive treatments;
he also could not tolerate psychotropic medication
because of adverse side effects. His depression
persisted and he began to show evidence of an
underlying dementia as well as intermittent pe-
riods of delirium. There were numerous episodes
of dehydration as his oral intake decreased. These
episodes would exacerbate his renal insufficiency
and he would become more delirious or obtunded.
Intermittent intravenous hydration would result
in periods of greater coherence and orientation.
However, hydration had to be monitored closely
because of the risk of congestive heart failure.

In his fifth month of hospitalization his close
friends brought in a copy of a living will he had
completed prior to his hospitalization. During a
period of lucidity, when he was deemed to have
capacity for medical decision-making, he reaf-
firmed the items enumerated in the living will,
specifically that in the event he faced a hopeless
prognosis, no cardiopulmonary resuscitation, no
intubation, no dialysis, no antibiotics, and no
tube feedings be done.

At one point, he participated in his own dis-
charge planning and agreed to go to a nursing
home rather than return home. He felt that the
burden of his medical and psychiatric care com-
bined with his wife's custodial home care could
not be provided effectively in their small apart-
ment. He adamantly did not want his wife to go to
a nursing home because her situation was stable
with a home health aide in their apartment.

Two weeks prior to his scheduled discharge to
a nursing home, he experienced a myocardial in-
farction and was transferred to an acute-care
medical unit. Cardiac evaluation subsequently
revealed that he had severe congestive heart fail-
ure and end-stage ischemic cardiomyopathy. He
lapsed into a stuporous state, unresponsive to
supplemental oxygen and diuretic therapy. He
did not eat, required total nursing care, and only
occasionally opened his eyes in response to his
name.

Both medical and nursing staff expressed
concern about whether an acute-care hospital
unit was the appropriate place for him in light of
his poor prognosis and because no acute medical
care was being provided. The nursing home that
had previously accepted him now refused him ad-
mission, stating that his medical condition was
not stable enough for him to be transferred at

Vol. 60 No. 1

THE PHYSICIAN'S EXPERIENCE— GOLDFARB ET AL

that time. His friends visited often, repeatedly
stating that a vegetative state was precisely the
situation he had wished to avoid by writing his
living will. His wishes, as outlined in his living
will, were respected, and no tube feedings or par-
enteral nutrition were initiated. He died ten days
later of intractable congestive heart failure.
I Questions Posed by Living Wills and Ad-
I vance Directives. This case highlights some of the
key questions concerning the doctor-patient rela-
tionship. Physicians and caregivers often find it
easy to go along with patients' and families'
wishes when they conform to their own. In con-
trast, they become resistant and uncomfortable
when these wishes diverge from what caregivers
think is reasonable or clash with caregivers' own
values.

Once a physician enters into discussions of
advance directives with a patient, does this com-
mit the physicians to actively perform the pa-
tient's advance directives? Does it ensure others'
compliance? For instance, in this case, is it the
physician's responsibility to withhold food and
parenteral nutrition from the patient? What
would have been the physician's responsibility if,
during one of the patient's lucid moments, he ex-
pressed his view that his intractable depression
had become unbearable and that he wanted assis-
tance in ending his life? How should a physician
assess a patient's decision-making capacity when
he or she is suffering from a psychiatric illness?
As obtaining advance directives becomes a stan-
dard legal and a standard procedural part of the
doctor-patient relationship, physicians and care-
givers must gain a greater understanding not
only of what these directives mean to patients,
but also of the responsibilities and obligations
these directives entail.

Child Patients
Audrey Schwersenz

The doctor-patient relationship has changed
quite a bit from the days of the omnipotent and
omniscient Ben Casey. Decades ago, living will
and do-not-resuscitate orders were barely dis-
cussed; many patients, perhaps most, quietly
waited for their physicians to make personal and

I important decisions for them. Whatever has
heightened patients' awareness of their own
rights and autonomy, whether it be better educa-

I tion, more liberal politics, or the ubiquitous mal-
practice lawyer, it has changed the way medicine
is practiced today.

The child patient, however, is a minor who is
usually not allowed self-determination of his or
her medical needs. A pediatrician is often placed

in the role of surrogate for the young patient
when a parent is absent or unfit. I am a pediatric
GI fellow who has been working in Mount Sinai's
liver transplant program for almost two years,
and I was recently involved in a case in which the
doctors and not the patient were responsible for
determining whether a child was going to have
any future at all.

A 12-year-old boy was referred to my service
for a liver transplant in 1990. He had been born
with extrahepatic biliary atresia which was suc-
cessfully treated with a Kasai operation in in-
fancy. About a year before we at the service met
him, he developed progressive jaundice, ascites,
coagulopathy, fatigue, and hypersplenism second-
ary to cirrhosis.

The nonmedical issues in his case were less
straightforward. He had a history of developmen-
tal delay and learning disability requiring special
education. A year before, his stepfather was
killed in street violence. The child became truant
and difficult to handle, and he did not fare better
when he was moved down south to live with his
grandmother. Eventually he returned to New
York, where he was placed in a group home for
children with behavioral problems and other dis-
abilities. His mother continued to maintain close
contact with him, and the ultimate plan was for
him to return home.

He was admitted for a pretransplant evalua-
tion that lasted over a week. I found him to be a
soft-spoken, good-natured boy who had difficulty
expressing himself verbally in an intelligible
way. Our studies were extensive and required
multiple invasive and noninvasive procedures
and medical consultations, which he tolerated
without complaints. Psychological testing and
evaluation are part of the study, and our child
psychologist felt that he was a good candidate for
transplantation.

Our evaluation required that he have a CT
scan. Orders were that he take nothing by mouth
overnight, and the scan was delayed until the af-
ternoon. In the radiology suite there were further
delays. He complained that he was hungry and
then left the procedure. He was eventually
brought back by security. My guess is that he was
just going to the cafeteria. Nevertheless, Mount
Sinai requires that such a patient be observed
one-to-one and receive an emergency psychiatric
consultation. The psychiatrist saw the boy in his
most stressed state and documented that his im-
pulsive behavior and borderline mental retarda-
tion made him a poor risk for a transplant.

I now had two trained psychiatric profession-
als with opposing opinions regarding the boy's
candidacy. I called his group home in order to get

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

a third opinion from a psychiatrist who had
known him for over a year. She told me that he
was not mentally retarded, but that he did have
an impulse-control disorder and expressive-lan-
guage disability. I also spoke to his pediatrician,
who confirmed that he was very cooperative ex-
cept when he was hungry. I had no trouble sym-
pathizing.

I had to present the boy's case to our recipient
review committee to establish whether or not he
was a candidate. I reviewed the facts of the case
for and against transplantation.

FOR:

• a transplant was medically indicated

• he wanted one, and his mother showed no am-
bivalence

• he was not profoundly retarded; in fact his dis-
abilities could be attributable to chronic liver
disease

• being a child, his future was totally unpredict-
able

AGAINST:

• he had an unstable home environment

• there was no way to be sure whether he could
ever take responsibility for himself

• his impulsive behavior might make it impossi-
ble for us to provide adequate care either in the
hospital or at home

To be honest, I never created such a formal list. In
the end, my own opinion was shaped less by those
facts and more by my instinctive feelings about
the boy based on my many interactions with him.

As objective as the recipient review commit-
tee aims to be, the influence of the way in which
the patient's case is presented cannot be underes-
timated. There were no rules that said I had to
present all the reports of the various psychiatrists
and psychologists. I knew that it might hurt the
child's chances of acceptance to discuss the views
of the emergency consultant in particular. But
my own ethical beliefs led me to present all the
opposing views as fairly as possible, and I asked
that all people involved in the case be present to
represent themselves at the meeting. Needless to
say, there was much hesitance to list him as a
candidate for transplant.

The debate continued for several weeks, and
the decision was continually postponed. I was vo-
cal about my own opinions. For example, I asked
the committee whether a reformed alcoholic is
more likely to be compliant than a child with a
learning disability. Why not consider the boy

more compliant than most, especially since he
lived in a structured group home? Should poten-
tial compliance be used as a criterion for child
recipients, since we should expect total noncom-
pliance during that awful period of time known as
adolescence?

I encouraged everyone to meet with and
speak to the boy before coming to a conclusion. In
the end, he "won" himself a transplant by charm-
ing Dr. Miller with his good nature and his ca-
pacity for sophisticated analogy. He compared his
transplanted liver to a race car engine, and un-
derstood that maintaining his liver would also re-
quire frequent check-ups, many tests, and tune-
ups and taking medicines every day (actually,
comparing cyclosporine to motor oil isn't that far
off). He eventually did receive a transplantation
and — luckily — he was a model patient. We have
not withheld any meals since his second postop-
erative day, and he has happily gained a whop-
ping 40 lbs. since the surgery.

The Child's Physician: Protective, Parental,
(Occasionally) a Pushover. My discussion here is
not on the ethics of transplantation, but rather on
my role as physician to my patient. I think that
the doctor-patient relationship in pediatrics is
unique. My motivation to advocate for the boy's
transplant was in part instinctive. Pediatrics al-
lows you to have those instinctive feelings and
close interactions with patients. You would not
feel uncomfortable hugging a child, or playing
with a child, or even acting somewhat childlike, if
it allowed you to communicate better, even
though these actions might be considered unpro-
fessional in adult medicine. Emotional distance is
harder to maintain with a needy and affectionate
toddler. It's harder to hate in pediatrics, since the
children are usually innocent bystanders, or oc-
casionally victims, of their parents' brutality, ir-
responsibility or self-destructive behavior. In the
end, you lose some of your objectivity, but never
your empathy. If being a pediatrician could be
summed up in just three words, they would be:
protective, parental, and (occasionally) a push-
over.

My responsibility as a child's doctor is to ad-
vocate for him or her what would be the best
course of action medically, keeping in mind issues
of quality of life. This is often colored by the pa-
tient's and parents' desires and, rarely, by my
own moral instincts. In choosing a treatment
plan, I do not believe that it is my role to deter-
mine what the burdens to society are, either fi-
nancially or socially, and I gladly relinquish that
responsibility to the ethicists.

The Doctor-Patient Relationship:

Panel Discussion

Moderator: Rosamond Rhodes, Ph.D.

Rosamond Rhodes: Let me briefly review where
we are at this point. Four philosophers and one
physician outline their views of what the doctor-
patient relationship actually is. Prof Bernard
Baumrin argues that the doctor-patient relation-
ship should appropriately be seen in terms of a
consumer contract, like buying tuna fish in the
supermarket. Prof. John Ladd has argued for the
ethics of virtue, where we take seriously our re-
sponsibility of other-regardingness. Prof White
argues for adopting a view of the doctor-patient
relationship in terms of a fiduciary contract that
pays a lot of attention to trust and promise. Dr.
Daniel A. Moros argues for a view of the doctor as
the trusted advisor to the patient, who is required
to use rationality and knowledge of science in do-
ing the right thing. And Prof Laurence McCul-
lough argues for seeing the problem of the doc-
tor's role in terms of finding the middle ground
between the duties of self-sacrifice and legitimate
self-interest. Then four physicians present cases.
Do the theoretical approaches help us answer the
questions raised by the doctors about what they
ought to do?

To begin, I ask Dr. Aufses, Ruth Ravich, and
Dr. Gorlin if they have any comments.
Arthur H. Aufses Jr. Speaking not as a surgeon,
but as a patient who had his first operation at The
Mount Sinai Hospital in 1930, I do remember
something about it, but I don't know what the
contractual relationship between the surgeon and
my father was. They were colleagues, since my

Adapted from a panel discussion at the conference "Whose
Contract Is It Anyway? Examining the Doctor-Patient Rela-
tionship" at the Mount Sinai Medical Center on March 8,
1991. Address reprint requests to the moderator, Director of
Bioethics Education, Mount Sinai School of Medicine, Box
1193, One Gustave L. Levy Place, New York, NY 10029.

father was a surgeon here. I do remember very
clearly having been operated on again at Mount
Sinai Hospital in 1943. I do remember that at
that time, there were no such things as consent
forms. If you were admitted to the hospital as a
patient and your doctor was a surgeon, it was as-
sumed that you were going to have an operation,
and you went ahead and had that operation, and
you either did well or you didn't do well. Dr.
Baumrin says that most contracts are implicit;
many are promises; and only a few are written.
And in those days, none of them were written.
But then he says, the patient is a purchaser of
services, the doctor is an offerer. That really stuck
in my craw, and I asked myself, how did we get
from point A, 1930, 1943, no written contracts, to
point B, where everybody is an adversary of ev-
erybody else? And clearly I think the answer lies
in a multitude of factors. I think the patients are
in part responsible. I have no doubt that the med-
ical profession itself is responsible in part. And I
also think that there are outside forces, third par-
ties— insurance companies, and lawyers — leading
to this change in the societal view of the doctor-
patient relationship.

Rhodes: Aside from being a model patient, Dr.
Aufses is also our chairman of surgery. Did you
have any questions that you wanted to ask of any
of the speakers?

Aufses: Dr. Goldfarb raised some interesting
questions. She asked whether doctors should use
their own biases and whether patients under-
stand that bias comes into play. I would say that
physicians, but especially surgeons, always use
biases in one way or another, although we try to
keep them under wraps as much as we can. And I
think all of us who deal with patients would agree
with you that patients only understand, or only
hear, what they want to hear. And by and large,
they do not want to hear things that they don't

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

want to hear, and they don't understand. It is to
your credit that in dealing with your patients you
enable them to understand what you're saying.
But how do you know if a patient understands
what you're saying?

Alisan Goldfarb: I can't simply use the fact that
they came to the same conclusion that I did as
evidence that they understood. Many patients
will clearly ask questions that signal that they
understand. Many, in fact, come with journal ar-
ticles with notations on the side, and they ask
sophisticated questions. And yet sometimes it is
clear to me that, although they are saying the
right words, they haven't put together some of the
pieces. That sensitivity may only come with years
of seeing patients. And that is where my moral
dilemma comes in. Can I trust my biases — can I
say to myself, my bias comes from experience? Or
are my biases a reflection of other baggage that
comes with me as an individual? It's clear to me
that my colleagues who go to the same meetings
and read the same journal articles sometimes
come to different conclusions. As physicians,
making decisions for individual patients, are we
dealing with them as individuals, or are we deal-
ing with them as members of a group? Have we
brought into the account the patient's personal-
ity? Particularly in what I do, there are many
decisions that may not be medically ideal, and yet
are correct for a particular patient. How much
time does a physician devote to a patient to try to
figure out what is right for an individual?
John Ladd: "Bias" is a dirty word and so I think
you are downgrading what you're doing by calling
it making use of your "bias." I think that to make
a judgment, which might be called a value judg-
ment or assessment, on the knowledge that you
have, is not bias.

Aufses: There's an expression that good judgment
comes from bad experiences, which come from bad
judgment.

W. D. White: I think your analysis of this is as-
tute, and worth reexamining. The issues you're
dealing with, particularly in breast cancer, are
complicated. There are the communication prob-
lems, and the disagreement amongst the experts
about what is optimal treatment, the costs and
benefits of the various options, and the fact that
the options clearly relate to nonmedical factors
like lifestyles and the values of the woman in-
volved. There are also the paradigms which phy-
sicians bring to problems rooted in their histories
and their experiences. These elements go far be-
yond any formal declarations of the informed-con-
sent rules. They take you toward the establish-
ment of a humane medicine. And that is

essentially the direction of the Jay Katz analysis
in The Silent World of Doctor and Patient, a book
that I recommend.

Laurence McCuUough: I think we can say some-
thing about what it means to understand. And
actually it points to an interesting research
agenda in clinical medicine. Some people are dis-
tinguishing two kinds of understanding. One is
cognitive — I've got to know what my problem is,
I've got to know what will happen if you don't do
anything for me. And I have to have some sense of
what might happen if you do this or that. It's the
notion that I have certain possibilities and I can
trace their implications reliably into the future.
That kind of understanding is a kind of a mastery
of information. The other, which is often over-
looked, is evaluative understanding.

Of these alternatives that you present to me
as the patient, say, to manage breast cancer,
which one makes the most sense in terms of
what's important to me for my health, how I want
to appear to others, my relationship with my hus-
band, others, and the like? All of that has to be
taken into account.

Since the informed-consent process only reg-
ulates what physicians are obligated to do, in
other words, to give information, it hasn't looked
at the other side, at what the patients have to do.
They've got to cognitively and evaluatively un-
derstand, and they need help.

The assumption is that we can just do this
routinely, but we can't. These are very compli-
cated pieces of information. Simple things, like
asking the patient, "What's important to you in
managing your breast cancer?" and getting the
patient to say, out loud for herself, maybe for the
first time, what that would be, sets a basis for
some of these discussions. It's not as mysterious
as it's been left. Appreciating the complexity
opens up a whole research agenda. What would
cognitive understanding be? What would clini-
cally count to show it has occurred? What is eval-
uative understanding? What would count clini-
cally as that occurring? These are empirical
questions we could answer.
Daniel A. Mores: To go back to my own concern,
the role of the physician as an applied scientist,
and the difficulty of applying science to human
life, I thought that Dr. Goldfarb illustrated what
a struggle that is, because there were things she
definitely knew, and things that she was less cer-
tain about, and then she was also confronting the
ambiguity of applying this sometimes more and
sometimes less certain knowledge to an individ-
ual patient.

She made an intriguing comment about what

Vol. 60 No. 1

DISCUSSION

^ she might have done differently. I believe this
process of searching through one's experience,
and critically evaluating oneself, is a central fea-
ture of good applied science and good applied eth-
ics. I wonder. Dr. Goldfarb, would you do the same
thing again? Do you think, in retrospect, that
your own conclusions were correct?
Goldfarb: Would I have done the same thing had
I the opportunity? I have, unfortunately, had a
similar patient since. I treated her differently.
And I unfortunately had the bitter pill to swal-
low. After I made my suggestion she sought care
with another physician.

White: But the burden of joint rather than uni-
lateral decision-making is the real issue here,
isn't it?

Goldfarb: I think that by the nature of my prac-
tice, there is no question but that all decision-
making is shared. The only decision that is not
shared is that every abnormality gets a biopsy.
Once we have biopsy information, then what to do
with that information is always shared. I have
been in practice for 11 years, and in all those 11
years only two patients have sat across my desk
and said, with a straight face, "Whatever you say,
doctor."

Richard Gorlin: One of the issues I am concerned
with is the tension between physician paternal-
ism and patient autonomy. If you are going to
participate as a patient advocate, you have to
take on a certain paternalistic role. You have to
do so to some degree, because otherwise, you're
going to sit there and simply say to the patient,
here are the four options. I'll present them to you
as best I can. I'll give you an objective evaluation
of the data as best I can. It's a Chinese menu, you
pick. And in some sense, this approach meets a
certain obligation. But as somebody who presum-
ably understands the patient, a physician has a
role as a counselor, an advisor, an advocate. All of
this depends on your assessment of the person's
ability to understand. How about the many pa-
tients who don't have that comprehension be-
cause of borderline mental status or a language
barrier that's difficult to get past? Then there's
the issue of the snapshot: you see a patient who is
called on to make life-threatening or life-trans-
forming decisions — how often is that done at a
single session? Do you tell the patient that this is
the big one? One of my cases might involve heart
surgery. Should the person have that operation or
not? Should the individual have a week for con-
templation and a chance to assess the options?
What is it I really want to work out with them?
How does hostility toward a patient bias the in-
teraction or impair the decision-making process?

That has always concerned me. It can almost be
punitive.

Rhodes: I have a question for all of you. The issue
came up particularly in Dr. Kalb's and Dr. Gold-
farb's case. In Dr. Kalb's case there is an attend-
ing physician, and there are also the physicians
working in the ICU. And in Dr. Goldfarb's case,
there was the gynecologist who wanted to do a
procedure that Dr. Goldfarb thought was inappro-
priate. When another physician is doing some-
thing inappropriate, or when you're involved in a
situation of multiple physicians working together
on a case, who is the doctor for the patient? What
is the relationship of these many doctors to the
patients?

Thomas H. Kalb: In my experience, each case pro-
vides its own answer. And the process that occurs
in the intensive care unit and in a lot of medical
settings is not dissimilar to the process that was
described in the pretransplant meetings, in that
different professionals advocate different posi-
tions. Decisions are made by force of science, by
the force of persuasion, and by the individual nu-
ances of the case. According to the ICU model, an
attending physician in the ICU is functionally
running the show. However, the ICU attending
has to deal with institutional regulations and the
reality of the fact that each patient has an admit-
ting attending physician, and that a balance
comes into play with each patient that depends on
that individual physician. Sometimes this is all
explicitly stated and sometimes it's only implicit.
Ruth Ravich: Patients are not told about the
rules of the hospital, so they don't know what
they need to understand. For example, when a
patient goes into the ICU, the attending physi-
cian should explain the rules to the patient and
the family. It needs to be said that theoretically
I'm still your doctor, but what's going to happen is
going to happen there, and I'm not the person who
is going to be consulted about it. We really don't
pay enough attention to telling the patients the
formal rules and the informal rules. And in this
case, there's a difference between the formal rule
and the informal rule, and it makes it even more
difficult.

Rhodes: Dr. Kalb, let me make my question a bit
more pointed. The patient is always your patient,
you're always the doctor. Is it ever appropriate for
doctors to disagree, to challenge the primary at-
tending?

Kalb: That's the way it always happens. That's
the norm.

Rhodes: Do you let the patients know that you
disagree?

Gorlin: That is exactly what I wanted to get to.

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

There has to be an accountable primary physi-
cian. That responsibility can move around. But
then it ought to be explicitly stated by the pri-
mary physician to the patient, that the baton has
been passed to another individual who becomes
the primary advocate of the patient. I believe that
one of the greatest sources of discontent and con-
fusion for the patient is when each successive con-
sultant comes in and says, "Well, I think you
ought to do this" instead of saying, "I will talk
with your doctor, and we will arrive at a final
recommendation that will be presented to you for
discussion." I think that as physicians we often
fail in this responsibility. There is always the
temptation to act as number one.
Kalb: In the case that I described, I think the
transfer had been made along with that state-
ment. There was an explicit understanding that
authority had shifted, and, in fact, the patient
was in a different room, had met different doctors,
had spoken with them. But when push came to
shove, in this particular case, the transfer of au-
thority broke down on an emotional level. So
you're right. What the patient believes and what
the doctors are doing in performing their duties in
dealing with the ambiguities involved in a case
are two different things. And I agree they should
be considered as discrete and separate.
Ladd: I have only one remark, and that not as a
philosopher but as a past patient. I was in the
hospital a few years ago. You sit in the hospital —
and I'm a fairly intelligent person, and they know
I'm a professor, but these guys come in and start
telling me, "I'm Dr. So-and-So, and you're going
to go in to such-and-such tomorrow." And then
five minutes later somebody else comes in and
says, "I'm Dr. So-and-So, and this is going to hap-
pen to you." After a while, I said to my own doc-
tor, "Who is this and that Doctor So-and-So?" He
said, "He's not a doctor at all, he's a medical stu-
dent. And that is somebody who's just helping us
out." What happens is that, as a patient, you don't
know what's going on. You don't know who's in
charge and what's going to happen. And I think
this is purely a matter of hospital mores which
could easily be solved just by telling people what's
going on.

Aufses: Well, I think it's a lot easier to deal with
this particular problem on a surgical service, be-
cause not only ethically but legally, the individ-
ual who contracts to do the surgery is responsible.
McCullough: May I talk out of school? I'm not
going to describe your institution, but one in
which I work in Houston, which has an open adult
ICU. These problems arise all the time. Perhaps
the attending of record can't be reached. Now if

I'm the family physician and I were to send Ms.
Ravich to Dr. Aufses, I would not be in charge of
the surgery. Dr. Aufses would be in charge of the
surgery. And I would not presume to tell him how
to practice his specialty. Yet the presumption is
often that the generalist or someone from another
subspecialty can tell the critical-care physicians
how to practice their particular specialty.

I asked my physician colleagues, why is your
ICU open? The bottom line is money. The attend-
ing physicians of record refused to let the ICU
become a closed ICU, because they priced out the
impact on their billings for the patients, and they
didn't like what they saw. It seems to me that
patients should understand that a particular pol-
icy is driven by what I think should be labelled as
mere economic self-interest on the part of physi-
cians.

Bernard Baumrin: I want to go back to tuna fish.
Go with me, for just a moment, a thousand years
into the future, when instead of our present mas-
sive ignorance we've actually got almost every-
thing figured out. In that kind of world, assuming
there still are physicians and you went to one
with a complaint, the physician would simply say,
"That's wrong, and the way to fix that is such-
and-such." And if the patient said, "Well, no, I
don't want to fix it that way," the physician would
simply throw the patient out. There wouldn't be
an issue any more. Having the knowledge, the
physician would then address himself or herself
to fixing the thing up with that knowledge. If
someone wanted to substitute their knowledge for
physician knowledge, the physician would just
tell them, "Go home and do it yourself, because
I'm not going to assist in applying nonscience to
something that has a scientific solution."

Back to tuna fish for a moment. What I was
suggesting is, the patient comes to buy something
which is, they hope, just like a can of tuna fish.
Some day, we hope, getting medical treatment
will be like buying a can of tuna fish. And when it
is, informed consent will be irrelevant; patient
autonomy will be irrelevant; hearts and flowers,
warmth and feeling, all this stuff will be totally
irrelevant. But in our present state of ignorance,
which is, after all, a lot better than it used to be,
we still have need of these things. I don't want to
be cruel and stupid and insensitive, but it's just
bedside manner. It's nice to befriend patients, to
help them through difficult periods, to help them
make decisions about the future disposition of
their affairs and the various kinds of therapies
that might be employed. But if we really knew
exactly what would work, they wouldn't be par-
ticipating in the decisions and it wouldn't really

Vol. 60 No. 1

DISCUSSION

matter how warm or friendly or kind or well-
meaning the physician was.
Aufses: Dr. Rhodes, you asked the question about
what happens when doctors disagree, and should
patients know that? And Ms. Ravich whispered
the answer right away: "Patients always know
it." It's clear, and Dr. Goldfarb gave you the ex-
ample. The patient came to her for a second opin-
ion or a third opinion, and then goes for another
opinion. Patients know that doctors disagree.

Prof. Baumrin went a thousand years ahead.
I would go in the other direction. All you have to
do is go back to 1905, and from 1905 to 1960.
Every doctor in the world knew that the only
treatment for breast cancer was radical mastec-
tomy. That was it, there was no other treatment.
You could give the patient radiotherapy, but the
treatment was radical mastectomy. And one of
the problems that the breast surgeons now face is,
they're doing lesser and lesser operations, and
rightly so, they're beginning to ask themselves
questions. Are we going too far in the other direc-
tion? And is this what medicine is all about?
We're not an exact science. We can cure appendi-
citis. We can take out the appendix and nobody
gets appendicitis again. But a lot of other things
are in a fairly gray zone. And I think that when
Dr. Goldfarb asks whether she did the right
thing, the answer is of course she did, at that
moment in time. But there may be a different
answer tomorrow.

Moros: I want to address comments to Dr. McCul-
lough and Dr. Ladd because their previous com-
ments go to the heart of what I feel has become an
increasingly powerful misconceptualization of
what medicine is, and this misconceptualization
produces negative results. Dr. McCullough sug-
gests that the structure of ICU practice is a re-
flection of financial interests. I grant that there
may be some truth to that. But we have a com-
plex, modern, somewhat scientific system that
still runs by rules and traditions that are one,
two, and three hundred years old, including the
original division of subspecialties and the struc-
ture of medical training. I see no sign that all
of the work of medical ethicists, all of the in-
volvement of legislatures and advocacy groups,
has in any way addressed any of the problems
of the inherent disorganization of medical care.
In fact, I think they've been slowly making it
worse.

Let me offer an illustration to support my
statement. Everybody here agrees that doctors
should talk to patients. For the last twenty years,
since I started in medical school, I have been
hearing that we are going to shift resources to

primary care, that we are going to pay doctors
more to talk to patients, and that we're going to
get doctors who talk to patients. But I know very
well that today, we pay a smaller percentage of
the medical budget for talking to patients, and for
caring for patients directly, than we did twenty
years ago.

So all the involvement of additional parties
trying to rationalize and modernize the structure
of medicine has failed. I'm concerned that some-
times doctors protect a financial prerogative, for
example by the way the ICU is structured. But
I'm more concerned about the fact that everybody
else who has gotten involved in restructuring
medicine may actually have made it worse.
McCullough: The history of funding medical care
gives a different picture. The present structure
reflects how Medicare and Medicare's payment
schedule has influenced the private paying sched-
ule. Medicare was created to provide health care
for all older Americans. But that's not the reality.
To get the bill passed. President Johnson sat
down with the American Medical Association and
the American Dental Association and the Amer-
ican Hospital Association and said, What do you
want? The American Dental Association said,
"Leave us alone." And he said, "Fine." The AMA
and the AHA said, "We want to be in control." He
said, "You have a deal. I'll give you a hospital-
based, physician-controlled reimbursement sys-
tem."

What do hospitals do to generate the reve-
nue? They do procedures. They don't do outpa-
tient primary care, talking to patients kinds of
things. And that dominant payment scheme, one
could argue, has had a much larger historical in-
fluence.

Moros: If I told patients that I wanted to bill them
for talking to me over the phone for thirty min-
utes, they would think I was crazy. So it's not just
the doctors who have developed this attitude that
you get paid only for specifically doing something.
It's a pervasive attitude throughout our society.
It's true that doctors are guilty of it as well, but it
doesn't reside solely with the AMA.

I think even Professor Ladd's comment about
his experience of lost control in the hospital re-
flects some misreading of the medical enterprise.
The same thing happens if you go into a law office
or if you try to deal with construction engineers.
Whenever you have to place your vital interests
in the hands of people with specialized knowl-
edge, where you don't have the time or energy to
acquire that specialized knowledge, you lose a
tremendous amount of control, which can be very
distressing. I can see no way to avoid it, although

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

I can see ways to ameliorate it. I certainly think
we all should be sensitive to what patients are
forced to experience.

Ladd: I rise to my defense on this matter. First, I
didn't want to know everything about my heart, I
wanted to know when I was going to go into the
operating room. And that's not a matter of exper-
tise, that's a matter of either responding, "I don't
know," or else somebody telling somebody else.
But to get five different answers within an hour
from people who purport to be doctors, who I know
are not doctors, is what I have in mind. Now that
doesn't really have anything to do with the larger
issue that you mentioned.

Rhodes: Going back to Dr. Leifer's case (and I
should point out that Dr. Leifer is one of Prof.
Ladd's former students), Dr. Leifer raised the
question of whether it is the doctor's responsibil-
ity to assess the patient's mental status and abil-
ity to make a judgment about their own care.
Leifer: The practice of medicine involves the day-
to-day, minute-by-minute assessment of patients'
capacities to decide about the appropriateness of
their care. We reflexively monitor what we think
of the decisions they are making. Is it in keeping
with their own set of values? It's an issue to be
stressed, especially at the end of life. As Dr. Gold-
farb said, the doctor-patient relationship requires
the doctor to make sure that patients understand
and are making decisions consistent with their
own values.

Goldfarb: Luckily, I don't have to deal with end-
of-life decisions. But I'm dealing with patients
making other decisions all of the time. I'm often
faced with the issue that a competent patient
with the capacity to make a choice who is clearly
a rational individual in all aspects of life is mak-
ing what is clearly an irrational, baseless deci-
sion. I can think of reasons why they came to the
irrational decision. Perhaps they are thinking
only of the present and not the future, or some-
thing else may be going on in their life, or they
fear a needle stick. Clearly, a psychiatrist is not
going to help me. Any psychiatric evaluation is
going to tell me that this is a competent, sane
person who is in a position to make all the deci-
sions, and yet I clearly see an irrational decision
being made.

McCuUough: Dr. Goldfarb, do you tell patients
that — that they've made an irrational decision?

Goldfarb: Yes.

McCullough: And how have they responded?
Goldfarb: Some of them say, "Yes, I know it's ir-
rational, but it's what I have to do." And some of
them tell me that I just don't understand them.
Rhodes: I have just one more question. In other
cases that have been described here, the doctor
seems have an obligation to the institution. Is
there a strict obligation only to the patients? In
Dr. Schwersenz's case, does she have an obliga-
tion to the committee, or only to her patients?
McCullough: It sound like she assumed that the
old motto, my patient comes first, is no longer the
sole story, that other parties can be affected by
what you do to and for your patients. In any
transplant of a scarce organ, where one person
gets it and someone else dies, you're not only al-
locating the chance to survive, you're allocating
death at the same time. I don't see how one could,
in those circumstances, realistically say, with any
faithfulness to the moral world in which you're
operating at that moment, "My patient comes
first." It sounds like Dr. Schwersenz was very
well aware that others must be taken into ac-
count equally as well as this child, and thus she
was objective. That is, she presented the evidence
that she had gathered, then forcefully advocated
for this particular patient.

Baumrin: She was a member of the committee,
and she had a duty to her colleagues to present
the information. On the other hand, she could cer-
tainly have attacked the information as un-
founded or not well evidenced, and whatever else
it might have been. Nevertheless, I would have
advised her to manipulate the committee as well
as she could have in her patient's best interest.
Gorlin: I think the physician is on a slippery slope
by ever being something beyond the patient's ad-
vocate-counselor. And that, I think, is the value of
the types of committees that you're describing.
This person should in fact be excused as other
than a presenter of the case.
Aufses: You forgot one little piece. Dr. Schwer-
senz said that the final decision was made when
Dr. Miller was persuaded that this young man
was an appropriate candidate. There is a commit-
tee of about 55 members. But when the chips
fell — going back to our discussion about who's in
charge in an ICU — the decision was made by the
person who has to do the transplant.

2. Transplantation

The Organ-Tissue Donation Process

Angelina Korsun, R.N., M.S.N. , M.P.A.

No MATTER HOW EFFECTIVE techniques are for
* preservation of organs and tissues, surgery, and
drugs, the success of transplantation depends on
the availability of organs and tissues. The history
of transplantation is long (Table 1), but the great-
est strides have been made in the last 25 years.

Only in the last few decades has organ and
tissue transplantation emerged as a successful
and clinically applicable therapy. As transplan-
tation expanded, the procurement process became
an increasingly important component. Thousands
of people require transplantation. At present, 266
transplant centers throughout the country per-
form kidney, liver, heart, pancreas, heart-lung,
lung, and bone marrow transplants and many
more perform corneal, bone, and tissue trans-
plants.

Renal transplant has the longest history of
the solid-organ transplants. Well over 80,000 kid-
ney allografts have been performed in the last 25
years. Extrarenal transplantation came into the
forefront in the 1980s, although a significant
number of patients were transplanted earlier. Ini-
tial survival numbers were poor, but pioneers in
transplantation continued to perfect the proce-
dures. With the introduction of cyclosporine, a po-
tent antirejection agent, graft survival improved
markedly and transplantation became more rou-
tine.

Survival rates for most transplants are in the
range of 75% and up. For many organs, it is rou-
tine to see survival statistics in the 85%-90%
range.

Adapted from the author's presentation at the conference
"Transplantation: Ethics and Organ Procurement" at the
Mount Sinai Medical Center on March 13, 1992. From the
Department of Surgery, Mount Sinai Medical Center. Address
reprint requests to the author at Box 1259, Mount Sinai Med-
ical Center, One Gustave L. Levy Place, New York, NY 10029.

Legislation

Significant legislation has been passed over
the years to promote and support organ and tissue
donation. Some of these legislative highlights
are:

1968 The Uniform Anatomical Gift Act

(adopted in all 50 states)
1972 PL 92-603, Medicare ESRD Network
1978 The Uniform Brain Death Act
1980 Uniform Determination of Death Act
1984 The National Organ Transplant Act
(PL 98-507): established Organ Pro-
curement and Transplant Network
(OPTN); established task force on organ
transplantation; prohibited purchase or
sale of organs or tissues
1986 PL 99-509, The Omnibus Budget Act
(OBRA), was the beginning of federal
required-request legislation: mandates
working relationship with OPO in all
hospitals receiving Medicare/Medicaid
reimbursement; mandates required re-
quest; mandates participation with
OPTN

1986 State-initiated required-request acts
(NYS 1986)

In New York State, as in other states, a re-
quired-request law mandates that each hospital
have a mechanism for identifying potential do-
nors and referring them to the appropriate pro-
curement agency. In addition, such a procedure is
required by Medicare as a prerequisite for reim-
bursement. In fact government is quite support-
ive of transplantation. The Medicare-supported
End-Stage Renal Disease network was created in
the 1960s, and Medicare now covers heart and
liver transplants in certain centers.

To put the magnitude of the donation prob-
lem into perspective: in January 1992 24,928 pa-
tients were waiting for solid organs, a significant

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

TABLE 1

Historical Highlights

1682

First reported bone transplant, Meekren

1969

Establishment of first network for organ distribution,

1881

First reported skin transplant

South Eastern Organ Procurement Foundation

1905

Introduction of corneal transplantation, Zirm

(SEOPF)

1938

Isolated organ perfusion, Carrel and Lindbergh

1980

Introduction of cyclosporin

1943

Hemodialysis machine developed, Kolff

1981

First successful heart-lung transplant, Reitz

1954

First successful kidney transplant in US, Murray

1977

United Network of Organ Sharing (UNOS) first

1962

First successful cadaveric kidney transplant

established as result of requests from non-SEOPF

1963

First successful liver transplant, Starzl

member centers

First successful lung transplant. Hardy

1984

Incorporation of UNOS

1967

First successful pancreas homograft, Lillehel

1988

Clinical use of Belzer's solution for organ

First successful heart transplant, Barnard

preservation

Adapted from ref. 1.

increase over the 13,766 waiting in February
1988 (Table 2). These totals do not reflect the ad-
ditional patients who may have been seeking a
bone marrow match, a corneal transplant, or skin
and bone transplants. The number of patients on
waiting lists has increased by 100% in the past 4
years, according to UNOS.

The unfortunate reality is that a large num-
ber of patients on waiting lists still die for lack of
an organ or a tissue match. Reported deaths of
patients on the solid-organ waiting list registered
with UNOS (all organs — kidney, liver, pancreas,
heart, heart-lung, lung) (4) were:

1988 1537

1989 1732

1990 2080

These totals do not include patients who have
died for lack of an appropriate bone-marrow
match.

The problem is not the lack of possible but of
actual donors. The number of cadaver donors in
1990 was 4,320; the approximate number of
deaths in this country is 2.2 mil/year; potential
organ donors are estimated by many experts at
14,000 (4). Over the past four years, the number

TABLE 2

United Network of Organ Sharing National Patient Waiting
List, Cadaver Organs Only

Organ awaited

2/1988

1/22/1992

Kidney

12,259

19,548

Liver

473

1,640

Pancreas

608

Heart

769

2,291

Heart-lung

159

154

Lung

687

TOTAL

13,766

24,928

Transplants (cadaver and

living related donor)

performed in 1991:

12,669

Adapted from ref 2.

of donors has remained fairly steady at about
4,000.

The United Network for
Organ Sharing

The United Network for Organ Sharing
(UNOS) was established in 1986. It is a volun-
tary, nonprofit group federally mandated by the
Organ Procurement Transplant Network to su-
pervise sharing, matching, and distribution of do-
nor organs. UNOS member hospitals are reim-
bursed under Medicare and Medicaid. UNOS'
main function is equitable organ distribution and
maintaining statistics on transplantation in the
United States. It uses single national lists of po-
tential recipients who are listed in order of point
score. Point allocation for kidneys is based on lo-
cal listing, time waiting, antigen matching, reac-
tive antibodies, medical urgency, and distance.
Point allocation for extrarenal organs is based on
local listing, donor size, blood type, time waiting,
medical urgency, and distance.

The Procurement Process. The crucial ele-
ment in the donation process is the call from the
hospital staff to an organ procurement organiza-
tion notifying them of a potential donor.

Nearly three quarters of the organ procure-
ment organizations (OPOs) are independent in-
corporated, nonprofit organizations. The remain-
ing quarter are hospital-based agencies (HOP As)
staffed by transplant donor coordinators, admin-
istrators, and specialists. They must be members
of the Organ Procurement Transplant Network
(OPTN) and must abide by its guidelines and pro-
cedures for brain death determination:

• Potential donor is identified by nurses, physi-
cians at participating hospital

• OPO is contacted

• OPO transplant coordinator comes to the hos-
pital to evaluate donor, assist in donor mainte-
nance, coordinate organ procurement

Vol. 60 No. 1

ORGAN AND TISSUE DONATION— KORSUN

• Brain death diagnosed by authorized physician
(OPO personnel may not participate)

• Consent for donation obtained; OPO personnel
may assist in completing consent forms and
counseling families

• Organs recovered with assistance of OPO sur-
gical team

• Organ maintenance and transport supervised
by OPO

• OPO reimburses the hospital on receipt of item-
ized bill and is reimbursed in turn by Medicare
or other payers

The clinical criteria for brain death are as follows:

• No response to external stimuli

• No reflex activity except of spinal cord origin

• No pupillary response to light

• No corneal reflex

• No eye movement with caloric testing or doll's-
eyes maneuver

• No gag reflex

• No cough reflex

• Apnea in the presence of adequate CO2 stimu-
lus

The following seven conditions are some of
the criteria used for identifying potential donors
of organs:

• Acute neurological or neurosurgical trauma

• Intracranial hemorrhage

• Gunshot wound to the head

• Primary brain tumor

• Drug overdose

• Metabolic disorder

• Cerebral anoxia

Organ donor cards and advance directives can be
utilized to indicate the wishes of the potential do-
nor. However, it is generally accepted that if fam-
ily members cannot come to agreement about do-
nating, even in the presence of a signed donor
card, the organs and tissue of that donor will not
be taken. That is why it is so important for people
who choose to sign a donor card to discuss their
wishes with family so that these wishes could be
respected at the time of death. In the absence of
an executed Proxy Form, family members may
give consent to the donation of the deceased's or-
gans or tissue in the following order of priority:
(1) spouse, (2) adult son or daughter, (3) parent,
(4) adult sibling, (5) legal guardian, (6) other au-
thorized individual.

Influences on the Organ Procurement Pro-
cess. Standards for donor selection have broad-
ened and become increasingly uniform. Not all
that long ago, transplant teams would not even

consider a donor older than 55 years of age. With
time and medical advances, we realize that phys-
iological age and not chronological age should be
the determining factor.

Brain death has become widely accepted as a
criterion for the determination of death and it has
become customary to employ the brain death
standard for the declaration of death.

Organ exchanges across the country have be-
come commonplace as preservation techniques
have improved. With the increased ability to pre-
serve certain organs, it is now routine to retrieve
organs in one state and transport them to another
area where the patient most needs that organ or
tissue or is the best match for it. This is facilitated
by UNOS and assures that a minimum number of
organs are wasted for lack of a recipient.

Techniques for organ removal have changed
from predominantly kidney and tissue only to
routine multiple organ retrieval. Emphasis has
been placed on the aggressive support of the po-
tential donor so that all of the organs may be in
optimal functional states. As skill has advanced
and more organs can be successfully trans-
planted, each donor must be supported to the full-
est to maximize the number of organs and
amount of tissue that can be retrieved from any
single donor.

Evolving Issues in
Organ Donation

Presumed Consent. In much of Europe the
rate of kidney procurement is substantially
higher per million than in the U.S. The U.S. av-
erage is about 26 kidneys/million of population,
whereas Austria (which has presumed consent)
reports 57.6 kidneys/million of population. In
spite of various legislative initiatives and in-
creased public awareness, the consent process
still remains an emotional issue. It is question-
able whether we will ever be able to have a pre-
sumed consent process in the U.S., even though
most major religions do not object to donation and
view it as an act of altruism.

Public Image and Myths. The media can be
friendly to transplantation, and many human in-
terest stories have been written about its bene-
fits. Yet there have been instances of television
productions giving the opposite message. An epi-
sode of "Law and Order" showed a man being
mugged in the park by a doctor and nurse, who
then remove his kidney in an apartment and re-
turn him to the park to die. This happens just
because a wealthy man wants this kidney for his
sick daughter. One "Knots Landing" episode sug-

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

gests that as long as you have enough money you
can buy your way to the top of the waiting list.

Many people fear that donation will result in
mutilation of the deceased, that an open casket
will not be feasible after organ harvesting, that
patients will not be cared for if hospital staff
know the patient has signed an organ donor card.
All of these beliefs are false, but such fears can
only be allayed if the correct information about
this issue is readily available and disseminated to
the public.

Financial Issues. Financial considerations
also have an impact on transplantation. Some in-
volve the sale of organs. In India, where organs
are, in fact, for sale, the legislature is now con-
sidering making this practice illegal. The issue in
the U.S. is whether we should provide financial
remuneration for donation. Who would pay; what
is a fair price? There are many questions and is-
sues that this provision would precipitate. Should
we pay for burial costs? Do you get a tax deduc-
tion for donation? Would such payments pressure
people who are economically indigent to obtain
some ready cash by selling organs?

Sadly, my office recently received two calls
from people offering to sell their organs for cash
since they were out of work and had heard that
you could get good money for organs. They were
quite disappointed when we informed them that
organ sales were illegal in the United States.

Access to Service. Transplantation is not a
service available only to the rich. In most states
Medicaid covers transplantation of kidneys,
hearts, liver, and bone marrow. Medicare covers
all of these, though with some restrictions.

The perception of inequality in access to
transplantation is a barrier to donation. Several
recent reports, including a report published by
the Office of the Inspector General, raise the point
of disparity among certain populations. It ap-
pears, from the data reviewed, that blacks wait
almost twice as long for an organ as whites. There
is substantial discussion about what these data
really mean, and many questions have been
raised about the data. How much is based on rac-
ism as opposed to medical issues? What are the
economic and social issues with regard to access?

Would this problem be alleviated if minority
donation increased? We know that the Bone Mar-

row Registry has predominantly white regis-
trants. A person who is Asian or African-Ameri-
can has a slim to nonexistent likelihood of finding
a match via the registry. This was one of the rea-
sons the government has granted additional mon-
ies to the Registry to actively promote minority
recruitment. In fact that has been quite success-
ful. The registry had until recently only about
100,000 to 150,000 registrants. They have just re-
ported reaching the 500,000 mark. However, only
about 30% of those are minorities. In order to
change this situation we need to promote greater
ethnic diversity in the donor pool so that all mem-
bers of society can benefit from transplantation.
There are several groups within the United Net-
work of Organ Sharing (UNOS) structure, as well
as groups not affiliated with UNOS, that are re-
viewing this information and developing pro-
grams directed at minority populations to remedy
the situation. Looking at the cultural differences
in the African- American community and the His-
panic community could enhance the consent rate.

Organs are in short supply. Who should re-
ceive them? Who should be given priority? Who
makes those life-and-death decisions? Although
dialysis is available for renal failure, and insulin
for diabetes, no substitutes are available for the
failed liver, heart, or diseased bone marrow. None
of these questions would be pressing if there was
no organ shortage.

These concerns must involve the joint effort
of healthcare professionals, legislators, patients,
and the public to make the choices that are re-
quired. Until this occurs we will continue to be
faced with many of the situations described by
other participants in this symposium.

My final comment, and the one I hope every
reader remembers most, is, "Don't take your or-
gans to Heaven — Heaven knows we need them
here!" To refer a donor in the NY area, call 1-800-
GIFT-4-NY.

References

1. Phillips M, ed. Organ Procurement, Preservation and Dis-

tribution in Transplantation. Richmond, VA: United
Network of Organ Sharing, 1991.

2. United Network of Organ Sharing data.

3. United Network of Organ Sharing.

4. United Network of Organ Sharing Update, 1991.

5. United Network of Organ Sharing Update, 1991.

Infants and Others Who Cannot Consent

to Donation

Thomas Tomlinson, Ph.D.

Most donors, of course, cannot consent at the
time of donation, because they are dead. Few
solid-organ donors have given explicit prior con-
sent to donation themselves. Typically, consent is
given for them by their surviving family mem-
bers.

But these are not the most problematic cases
of donation without consent, for two reasons.
First, most of these donors were competent adults
prior to the injury or illness that led to their
death, and so other people close to them had some
basis for judging what their wishes would have
been about organ donation. Second, since these
donors are dead, there is no longer any risk of
harming them, and so there is less at stake mor-
ally in guessing wrongly about their real prefer-
ences concerning organ donation.

The more difficult cases, therefore, are those
where these two conditions are not met: where
there is no prior history of individual values on
which to make a decision about organ donation;
and where the donor is still alive, so that the do-
nation imposes on them some risk of harm. There
are various sorts of persons who meet these con-
ditions. Retarded donors are one class, and there
have been some commentaries and court deci-
sions on such cases, notably on live kidney dona-
tion (1) and bone-marrow donation (2). But I want
to focus on another class of never-competent liv-
ing donor — infants.

The case I will use as a springboard involved
Abe and Mary Ayala, whose 17-year-old daughter
Anissa was diagnosed with myelogenous leuke-

Adapted from the author's presentation at the conference
"Transplantation: Ethics and Organ Procurement" at the
Mount Sinai Medical Center on March 13, 1992. From the
Department of Medical Ethics, Michigan State University.
Address reprint requests to the author, Professor of Medical
Ethics, at 1550 Ann Street, East Lansing, MI 48823.

mia. Without a bone-marrow transplant, odds
were 80%^90% that she would be dead within five
years. With a matched bone-marrow transplant,
her chance for cure would be close to 70%. But the
odds of finding a matched unrelated donor are a
dismal 1:20,000, and the match they lucked upon
refused to donate. Her parents then set about to
have another child, first requiring a reversal of
Abe's vasectomy, hoping to win the 1:4 odds that
the child would be a match for Anissa. In fact,
their new daughter, Marissa, was a match, and at
about six months of age, in June 1991, had bone
marrow removed for implantation in her sister
(3-5). According to recent reports (personal com-
munication, Elizabeth Quam, Bone Marrow Pro-
gram) the transplant was a success and both
Anissa and Marissa are doing fine.

Can Infants Already Born
Be Used As Donors?

Various arguments might be made against
what the Ayalas did. Some of these arguments
would be directed against the idea of deliberately
conceiving a child to use for organ donation. I will
consider these below. Other sorts of objections
would be directed against using any infant for
bone-marrow donation, whether or not the child
was conceived for this reason. I want to consider
these sorts of objections first, both because they
would apply to a much larger class of cases of
never-competent donors, and because the moral-
ity of conceiving children for use as donors will to
some degree turn on the morality of using infant
donors at all.

The Question of Excessive Risk. The first
concern to consider is whether using infants as
donors subjects them to excessive risk. For some
sorts of donation, risks may be appreciable. But
the risks of bone-marrow donation are exceed-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

ingly small, and well within the range of every-
day risks that parents are permitted to expose
their children to.

According to the National Bone Marrow Do-
nor Registry, out of over 5,000 donations per-
formed to date around the world, no bone-marrow
donor has suffered any long-term adverse effects
from donation. The risks are primarily those from
general anesthesia, which in adults average 1
death in 15,000 across all patients and facilities.
The risk of death from anesthesia of a healthy
adult in a top-flight facility would be something
closer to 1:40,000. This risk would likely be lower
in healthy infants, who are hardier and more re-
silient to anoxic insults to the brain (personal
communication, James Waun, M.D., Board-certi-
fied anesthesiologist).

One in 40,000 is about the additional risk of
death that would be created by taking the child
on a cross-country automobile trip to see
Grandma, or driving about town a comparable
distance during the year, as calculated from esti-
mated risk of automobile travel reported in Ac-
ceptable Risk (6). This is a risk that few of us
would consider excessive, even on the assumption
that few of these miles were traveled for the sake
of some vital interest of the child, but rather for
the convenience or interests of other family mem-
bers.

The Question of True Consent and the
Canon of Loyalty. A more serious objection
would contend that organ donation can only be
based on consent, which infants cannot give. The
argument here would be parallel to the reasons
Paul Ramsey gives against any use of children as
subjects in nontherapeutic medical experimenta-
tion:

Morally, no parent should consent . . . [to] nontherapeutic
. . . experimentation, [which] must be based on true consent
if it is to proceed as a human enterprise. ... No one else on
earth should decide to subject these people to investigations
having no relation to their own treatment. That is a canon
of loyalty to them (7).

Commenting on the Ayala case, George An-
nas put it more pithily: "Children are not medi-
cine for other people" (8).

The claim here is that organ donation, like
participation as a subject in human experimenta-
tion, involves imposing a harm or risk of harm on
one person (the donor or the subject) for the sake
of another person. If research subjects agree to
make this sacrifice, it is morally commendable,
but it is not a moral duty, even when the benefits
to be gained by others are substantial. The same
is true of organ donation, because it too is a sac-
rifice. To remove organs from infants or others is

in a morally suspect way to "sacrifice" them, in
the name of a utilitarian calculation which be-
trays the loyalty we owe to their sacred, individ-
ual personhood.

But just what is this "canon of loyalty" that
Ramsey speaks of? Certainly, we think that par-
ents have an obligation of "loyalty" to their chil-
dren, but what is the content of that obligation?
We can agree that at a minimum, it means not to
sacrifice one's child for the sake of someone else.
That is easy to accept, simply because the idea of
"sacrifice" is associated with killing, maiming, or
otherwise seriously harming the child. But as al-
ready explained, the risks imposed by bone-mar-
row donation are not of this magnitude, and so do
not demand "sacrifice" in any sense that is pa-
tently morally objectionable.

An alternative interpretation would be that
the demand of "loyalty" requires the duty not to
impose any risk on one's child not of its choosing.
But this alternative is clearly unacceptable, be-
cause on this interpretation parental consent to
most medical treatment would be a violation of
loyalty, and presumably it is not, as Ramsey him-
self allowed. On his view, the only time we are
justified in imposing risks on unconsenting in-
competents is when doing so is, on balance, to
their benefit. Indeed, the several court rulings
which have approved live organ or tissue dona-
tion from a retarded person for an ill sibling have
done so by arguing that it was in the retarded
person's best interest to donate and thereby in-
sure the continuing benefits of the sibling's com-
panionship, guardianship, and so on (1).

But why, or more accurately, when is consent
to treatment not a violation of loyalty? This is a
fair question, because there could be consents for
treatment of incompetent persons that were none-
theless disloyal — namely, when the choice made
for treatment was inconsistent with what we rea-
sonably believed the incompetent person would
have wanted. This would be an act disloyal to the
values that the person held dear, and which
formed part of their self-identity as persons.

By extension, I would argue, consent for
treatment of a never-competent child is loyal to
them only when it is based on reasonable pre-
sumptions about what the child's values and
goals would be relative to the risks of harm and
promises of benefit of the treatment. It is only on
the basis of such presumptions that we can make
claims about what is in the child's "best interest."
But if this is what permits us loyally to impose
the risks of treatment, the same approach should
also permit us to impose the risks of organ dona-
tion, when it would be reasonable to presume a

Vol. 60 No. 1

DONATION WITHOUT CONSENT— TOMLINSON

commitment of the child to the life its donation
would save, at very small risk to itself

It would be a paradoxical "canon of loyalty"
owed to persons if we were permitted to act only
on presumptions about the incompetent's self-in-
terested values, and not at all on presumptions
about the sort of moral values that make persons
beings who are specially owed a duty of loyalty. It
is not a violation of loyalty, but an expression of
it, when I conscientiously strive to make the de-
cisions my incompetent child would make as a
mature person, equipped with minimally decent
moral commitments to others (9).

One implication of this reasoning is that pa-
rental consent for organ donation to a sibling
would be more defensible than consent for dona-
tion to a stranger, because in the latter case there
would be less of a reasonable presumption that
the child would feel any commitment to saving
that person, as opposed to another family mem-
ber.

However, a third reason offered against pa-
rental consent for organ donation is that when
the recipient is another family member, parents
are in a conflict of interest as protectors of their
infant's welfare. Their love and desperate hope
for their other child might blind them to the real
magnitude of the harms their donor child would
suffer.

This is probably a good argument against re-
lying on parental consent as the sole and suffi-
cient warrant for proceeding with a transplanta-
tion. But the real danger it warns against can be
blunted by requiring an independent assessment
of whether the risks of transplantation are exces-
sively high, so high that we can no longer confi-
dently presume that the child donor herself would
want to take the risk for the sake of her sibling.
At a minimum, those who are coordinating the
donation and transplant should recognize their
responsibility to apply standards of reasonable
risk to the parents' decision to use a child as a
donor. I shall not attempt to work out the details
of how a completely adequate independent assess-
ment would be made, except to say that it cannot
consist simply in the concurrence of the trans-
plant team, who have their own predisposing
agenda (I thank a member of the conference au-
dience for reminding me of the transplant team's
conflict of interest.)

Can Children Be Conceived
For Use as Infant Donors?

Once it has been shown that using infants
already born as live organ donors is not in itself

morally objectionable, the main argument to be
made against conceiving children for that pur-
pose evaporates. Nevertheless, several other sorts
of objections have been made.

The Question of Children as Means. One ob-
jection has it that conceiving children as donors
uses them as a means, but children should be
brought into this world only for their own sake, as
ends in themselves: "The ideal reason for having
a child is associated with that child's own welfare,
to bring a child into being and to nurture it" (10).

Although Marissa Ayala was "used as a
means" to save her sister's life, this fact is not by
itself morally decisive, both because such use
posed little threat to Marissa's interests, and be-
cause we had, on the face of it, no cause to doubt
that her parents valued her for other reasons as
well. She was used as a means (as we all are), but
not merely as a means, which is the crucial mod-
ifier in the Kantian injunction to respect persons
as ends in themselves.

Except for the saintly (who thankfully have
few offspring in any event), most of us conceive
children out of a complex mixture of motives,
when the conception is "motivated" at all. Most of
these motives are self-interested, and few of them
are as altruistic as the Ayala's concern with sav-
ing Anissa's life — all with no palpably evil re-
sults. It is therefore naively pious to claim that
children "should be conceived for their own
worth." Children, once conceived, should then be
treated with due regard for their intrinsic worth.
But there are no reasons for thinking that
Marissa, or any other infant under similar cir-
cumstances, would not be so treated by her par-
ents. Except in the most extreme cases, where
there is only a single overarching or malevolent
motive, there is no natural connection between
motives for conceiving children, and later motives
and feelings which support nurturance and
love — a good thing, too, since the survival of the
species has probably depended on it.

The Question of Abortion Used Solely for
Donation. Another objection which has been
raised is that conceiving children as donors im-
plies acceptance of abortion for purposes of fetal
organ donation. After all, if someone with the Ay-
alas' motives discovered that the fetus they had
conceived was not a suitable match, they could
just as well decide to abort that one, hoping for
better luck with the next. Assuming that we
should condemn conception and abortion for the
purpose of transplanting fetal tissue, should we
not also condemn what the Ayalas did?

This is an obvious non sequitur. Even if we
should condemn all such abortions (an arguable

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

assumption itself), it would be because using the
fetus's organs requires killing the fetus, and by
killing the fetus, we would violate the respect
owed to the intrinsic value of fetal life. Using the
fetus as a means for organ donation by way of
abortion is incompatible with respecting it as an
end in itself Using Marissa as a means for ob-
taining bone marrow is not incompatible with re-
specting her as an end in herself Even the
staunch antiabortionist could consistently accept
the Ayalas' end, but reject using abortion as the
means to achieving it.

Conclusion

Using infants, and by extension, other never-
competent persons as live organ donors is not in
itself morally objectionable, nor is conceivmg
children for this purpose. When the risks to the
donor are very low, and the never-competent per-
son's loving concern for the threatened recipient
is reasonably presumed, the person is not illicitly

being used as mere means, nor sacrificed for the
good of others.

References

1. Strunk v. Strunk 445 SW2nd 145 (Ky Ct App 1969).

2. Matter of John Doe 481 N.Y.S.2nd 932 (A.D. 4 Dept.

1984).

3. Time is an enemy. Los Angeles Times, Feb. 24, 1990, B2.

4. Baby delivers hope and joy to family. Los Angeles Times,

April 7, 1990, B3.

5. Baby girl's bone marrow transplanted into sister. Los An-

geles Times, June 5, 1991, Al.

6. Fischhoff B, Lichtenstein S, Slovic P, et al. Acceptable

risk. Cambridge: Cambridge University Press, 1984:
81, Table 5.1,

7. Ramsey P. The patient as person. New Haven: Yale Uni-

versity Press, 1970, 14.

8. When having babies for "spare parts" makes sense. The

Washington Post, March 25, 1990, C-5.

9. For a somewhat different argument imputing moral val-

ues to never-competent children, also made as a reply to
Ramsey, see McCormick RA. Proxy consent in the ex-
perimental situation. Perspectives Biol Med 1974;
18(l);2-20.

10. Reported in Baby is conceived to save daughter. New York
Times, Feb. 17, 1990, Al.

Debatable Donors

When Can We Count Their Consent?

Rosamond Rhodes, Ph.D.
Abstract

Donor consent is the standard requirement for the acceptance of an organ or tissue dona-
tion from a living donor. Usually consent to donation is respected as the choice and act of
an autonomous agent. There are times, however, when the consent should be paternalis-
tically set aside. The problem for those who must make these decisions is determining
when the donor's autonomy should be respected and when paternalistic interference is
appropriate. This paper draws on the writings of Kant to develop an autonomy-preserving
criterion for determining where to draw the line. It then goes on to display how that
criterion could be applied to the case of an adolescent donor who agrees to donate a kidney
to his brother.

Whenever a person becomes a living donor, she
or he is subjected to some risk, pain, and mutila-
tion. Nevertheless, considering the good it does
for another, we usually accept the organs and re-
spect the donor's choice. There are times, how-
ever, when a willing donor is refused. When can
we count a donor's consent? When do we pater-
nalistically refuse to act on a donor's choice?
Why? To set the stage for understanding my an-
swer to these questions, I begin by presenting
some hypothetical situations and common intui-
tions.

If I wanted to cut off my healthy hand, or
even just a healthy fmger, I imagine that most
people would feel justified in interfering and stop-
ping me. I also imagine that I would not be able to
find a surgeon who would be willing to perform

Adapted from the author's presentation at the conference
"Transplantation: Ethics and Organ Procurement" at the
Mount Sinai Medical Center on March 13, 1992; draws on
author's research as participant in the National Endowment
for the Humanities 1990 summer seminar, "The Nature and
Value of Autonomy," led by Gerald Dworkin.

From the Office of the Director of Bioethical Education,
Mount Sinai School of Medicine. Address reprint requests to
the author, Director of Bioethics Education, Box 1193, Mount
Sinai Medical Center, One Gustave L. Levy Place, New York,
NY 10029.

the amputation for me, even if I were willing to
sign all of the necessary consent forms.

If I wanted to donate one of my healthy kid-
neys to my child who had kidney failure and
needed one, people would support my choice and
the kidney transplant team would be eager to co-
operate. Nevertheless, pressure from other family
members and society at large (which raises eye-
brows whenever parents survive a fire unscathed
while their offspring perish) could be so coercive
as to make a parent feel unable to withhold con-
sent. If, however, I wanted to donate my heart to
save my child, I am sure the heart transplant
team would refuse to cooperate. And if I wanted to
donate a kidney to each of my two children who
were in need, I imagine that the kidney trans-
plant team would not accept my consent as valid
and refuse to act on it.

If I was willing to be a bone-marrow donor for
a total stranger, my consent would be accepted
and I would be seen as a hero. On the other hand,
if I volunteered my 3-year-old child who, when
asked, said "yes," or presented my 7-year-old who
clearly articulated her willingness to donate mar-
row to save the dying stranger, it is not obvious
whether people would accept their agreement as
consent. If these children were agreeing to donate
a kidney instead, it would be even more likely

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

that people would refuse to count their consent.
Yet we might be inclined to accept the consent of
one child (or childlike individual) to donate a kid-
ney to the other, and we might even determine
that the child's consent was not necessary for the
moral acceptability of that donation; recent legal
cases have reached just such conclusions (1-3).

I suspect that this survey of responses pro-
duces a clear and uniform set of intuitions. When
death is a certainty for the heart donor, people
reject my consent. When death is only a remote
possibility, my consent to a nonmedically-indi-
cated amputation is refused, whereas my consent
to kidney or marrow donation, which may involve
no less a risk, or even a greater risk, is accepted.
So risk itself could not be the crucial consider-
ation.

When it comes to my children, their consent
is questionable, making our acceptance of their
donations uncertain. When one is needed as the
donor for the other, however, the quality of the
consent and even the need for consent seem irrel-
evant. So consent, in itself, could not be the sole
criterion for our decisions.

In the latter case of one child donating for the
other, the children's familial relation is signifi-
cant. But personal relationship is unnecessary for
the acceptability of my own marrow donation for
the dying stranger. Risk, consent, and relation-
ship all seem to be considerations in deciding
when to accept an organ donation, but how they
fit together and why is not clear.

We do need organ donations to save or im-
prove lives. Does this need allow us to accept the
consent of questionable donors (such as retarded
or demented relatives)? And why? Although some
of our intuitions about the hypothetical cases are
clear, the principle underlying the acceptability
of some consent is certainly not obvious. Why
does agreement sometimes have to be respected?
When can consent be set aside? If we accept the
prospective donor's consent, we respect their au-
tonomy. If we refuse to accept their consent out of
concern for their good, we are treating them pa-
ternalistically. Is such paternalism ever morally
acceptable?

It is common to see paternalistic action in
conflict with respect for autonomy, but there is a
way of reconciling the two, as the good parent
does. Emmanuel Kant, the nineteenth-century
philosopher who focused our attention on the pri-
macy of autonomy, wrestled with this problem of
justifiable paternalism. In what follows I adopt
the Kantian approach to the issue as it presents
itself in modern medical practice. First I lay out a
groundwork for understanding the dilemma by

presenting a brief sketch of autonomy and benef-
icence. These mere glosses of the concepts should
be sufficient for this discussion, and they are
meant to be neutral as to questions of their source
or generation. I then present a principle for as-
sessing when paternalistic intervention is appro-
priate. I conclude by illustrating how that adju-
dicating principle can be applied in resolving a
complex case.

A Brief Sketch of
Autonomy and Beneficence

Autonomy. Respect for autonomy is the most
fundamental of all (Kantian) principles. It de-
rives from the recognition that what distin-
guishes moral beings from other creatures and
what entitles persons to special treatment are (a)
the ability to conceive of moral principles or rules,
(b) the ability to choose their actions in terms of
conforming with moral rules, and (c) the ability to
limit their action to conform with those princi-
ples. Having these abilities, which amount to the
ability to be moral, is being autonomous. Auton-
omy, therefore, deserves ultimate respect because
it is taken as the ground for both moral treatment
and moral responsibility. Autonomous agents
each have their own conception of their own good
and guide their actions accordingly. Respect for
autonomy therefore requires that we allow people
to make their own choices.

Beneficence. Beneficence is also a basic prin-
ciple of (Kantian) ethics, derived from the under-
standing that anyone needing help would want
others to provide it. Since morality demands that
we treat others as we would wish to be treated
ourselves, we are obliged to treat others with be-
neficence.

A Principle for Assessing
Paternalistic Intervention

Paternalism. With these preliminary under-
standings of the principles of autonomy and be-
neficence, we can proceed to an examination of
paternalism. According to my understanding of
Kant, the problem of paternalism arises because
relationships among rational beings in the moral
world are governed by both the principle of mu-
tual love and the principle of respect.

The principle of mutual love gives us the
"maxim of benevolence," which, for Kant, entails
the "duty to love one's neighbor" or "to make the
ends of others (as long as they are not immoral)
my own" (4). The second principle, respect,
charges us to limit our self-interested behavior in
order to preserve the "dignity of humanity in an-

Vol. 60 No. 1

DEBATABLE DONORS' CONSENT— RHODES

other." It entails the strict "duty of free respect to
others," which is analogous to the duty "not to
encroach upon their rights." Whereas mutual love
constantly directs people "to approach one an-
other," respect directs them "to keep themselves
at a distance." The duty of benevolence seems to
require involvement in the lives of others; respect
seems to require that we leave others alone. The
problem for the autonomous agent — as for exam-
ple the health care provider — who governs action
according to these principles of duty is to deter-
mine when paternalistic action, as in action that
interferes with a patient's choice, satisfies the re-
quirements of both mutual love and respect.

Assessing Autonomy in Others. When con-
sidering paternalistic interference, a reason for
presuming that the other is capable of making an
autonomous choice is sufficient for attributing
autonomy. A sufficient reason for requiring that
another be respected as an autonomous agent
would be their demonstration of self-motivated
action, or voluntary behavior. Stated negatively,
and more to the point, only reasons for concluding
that the other is incapable of making an autono-
mous choice (the particular one in question) could
justify a particular paternalistic interference.

This standard presents a strong presumption
in favor of regarding others as autonomous be-
ings, capable of making rational choices relative
to their own goals. Even when another's choice
strikes us as a bad idea, foolhardy, or dangerous,
for the most part we must respect it and leave
others alone to take whatever risks they choose.
In Kant's framework, we are bound to "cast a veil
of philanthropy over the faults of others" (ref. 4,
466). Respect for the other as an end in himself
requires us to presume, whenever possible, that
his action is autonomous. We have "a duty to re-
spect man even in the logical use of his reason:
not to censure someone's errors under the name of
absurdity, inept judgment, and the like, but
rather to suppose that in such an inept judgment
there must be something true, and to seek it out"
(ref. 4, 463).

In other words, as far as possible, Kant de-
mands that we try to consider others' actions as if
they were freely done from an autonomous will.
Respect requires us to try to think of what an-
other does as something we too would find rea-
sonable if only we knew the facts of the situation
as we should imagine the other does.

Obviously, since paternalistic interference is
ideally applied before another's autonomy-threat-
ening act could be performed, in some cases the
other's choice could be the salient consideration.
We could decide that although there was no inde-

pendent reason (such as known intoxication or
psychosis) to suppose that the person lacked the
capacity for autonomy, if this particular choice
were enacted the person's future autonomy would
undoubtedly be compromised and, furthermore,
we might be unable, despite a sincere effort, to
conceive of a rational end that the imminent cho-
sen action might serve. In some cases (as Kant
might argue) an assessment of the other's capac-
ity for autonomous willing is the paramount con-
sideration: a person might be either incapable of
willing at all or incapable of reason.

In each situation, after finding the other's
choice inappropriate or capacity inadequate, the
paternalistic agent must ascertain whether or not
failing to interfere would make the other's future
autonomous action impossible. If so, the agent
would be morally bound to perform the paternal-
istic act of beneficence. On the other hand, when
there would be no irremediable effect on the oth-
er's autonomy, regardless of how foolish or unde-
sirable one might think the action, "[w]hat they
count as belonging to their happiness is left up to
them to decide" (ref. 4, 388).

The end of the other's continued existence as
an autonomous being, therefore, serves as the
limiting conception for paternalism:

• When autonomy can only be preserved by pa-
ternalistic action (for instance, stopping me
when I want to cut off my own hand), because it
preserves autonomy, the only thing which has
absolute worth, we must interfere.

• When a person's capacity for autonomous ac-
tion is intact and can be expected to remain so
(as when I want to donate marrow to a
stranger) paternalistic interference would not
be justified, since autonomy is not jeopardized.

• If, however, there is no possibility of preserving
autonomy (for example, if the requisite mental
status can never be restored, since the brain
has "liquified"), no duties qua "end in itself
can be owed to the other.

Tests for Appropriate Paternalism. Physical
welfare as well as psychological well-being (ref. 4,
393-394) count as needs which must be met for
continuing agency. Seriously compromising ei-
ther (but not merely making oneself somewhat
worse off or somewhat diminishing one's chance
for continued agency) would make future auton-
omous action impossible. These autonomy-pre-
serving justifications for paternalistic interfer-
ence explain why Kant sees "minors and the
mentally deranged" (ref. 4, 454) as broad classes
of people who are generally exempted from the
autonomy-respecting rule. Without autonomy-

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

preserving paternalistic interference, the future
autonomy of minors can be jeopardized. Without
such interference, the temporarily deranged
might never have their autonomy restored.

Besides referring to needs as a test in deter-
mining the appropriate domain for paternalistic
interference with another, we can also rely on a
model of hypothetical assent. In judging whether
an act of paternalism would be my duty, I would
invoke hypothetical assent and ask whether, as a
rational being, one could possibly assent to the
other's proposed action. Paternalistic interference
would be justified if the paternalistic agent were
to find that he could not "contain the end of this
action in himself"

To illustrate the employment of this model,
again consider some examples. Paternalism
would be justified when I wanted to cut off my
healthy hand. Because an agent could not imag-
ine any end which she or he might adopt that
would motivate such an action — that is, hypothet-
ically, could not conceive of himself or herself as-
senting— the agent must paternalistically act to
preserve autonomy and stop me from severing my
hand. Paternalism would not be justified for the
bone-marrow donor because an agent could imag-
ine a rational person assenting to accept some
pain, scarring, and inconvenience to save an-
other's life.

This hypothetical assent test differs from
other models for delimiting paternalism which
rely on hypothetical assent (5-7) in that it does
not ask the paternalistic agent to conceive what
she or he could not know — whether the benefi-
ciary would or would not assent, which is the test
invoked in "substituted judgment." Hypothetical
assent is assessed straightforwardly, on this view,
according to one's own conception of what would
gain the assent of any rational agent. (Concern
about the patient's idiosyncratic views, personal
history, previous statements, and so on, have lim-
ited relevance according to the rational agent per-
spective, whereas they would count a great deal
for those who base hypothetical assent on a deter-
mination of what the particular beneficiary
would have wanted. Furthermore, healthcare
workers must always be alert to the possibility
that their own personal history may influence
their view of rationality. It is the view of ratio-
nality, rather than the view of the patient's or
healthcare provider's idiosyncrasies, which may
justify and direct paternalistic interference.)

The need test and the hypothetical assent
test for the appropriateness of paternalism relate
in a typical Kantian fashion to the form of the

agent's willing rather than to external criteria of
the "beneficiary's" condition. Instead of requiring
proof of the other's lack of rationality or compe-
tence, the moral standard demands only a sincere
effort to test one's judgment by the autonomy-
preserving principle of paternalism.

Application

With the autonomy-preserving criterion for
paternalistic interference in mind, let us examine
the complicated case of a 17-year-old who is the
best organ match for his young brother with kid-
ney failure. Although a transplant from this sib-
ling would appear to be the optimal medical
choice, the moral acceptability of that option is
certainly not obvious. The consequences of taking
and not taking the kidney need to be considered,
and we also need to review the moral principles
bearing on the case. (A similar case was debated
for several weeks in 1991 at the weekly Mount
Sinai Kidney Transplant meetings chaired by Dr.
Lewis Burrows; see ref. 8.)

Case Report. The potential donor, David,
wants to donate a kidney to his brother, Ken. He
understands that the procedure presents some
risk for him and that after surgery he may no
longer be able to continue playing football. His
idols have all been football players and he now
plays on his high school team. Nevertheless, he
agrees to be a donor as soon as the option is men-
tioned, and he never displays any ambivalence.
He says, "I want to donate my kidney because
then I'll be a hero to my family." This close fam-
ily, of two parents and five older siblings, strongly
supports the 17-year-old's decision, especially af-
ter an older brother is found medically (anatom-
ically) unsuitable.

The parents and two of the older siblings
could also be medically acceptable donors. How-
ever, their organs are likely to be better grafts
(one haplotype matches) than a nonrelated cadav-
eric kidney, but they are less compatible than the
perfect organ match (haploidentical) that could be
provided by David, the adolescent brother. Stud-
ies have shown that in the short run there is little
difference in the survivability of organs from dif-
ferent classes of donors. After several years, how-
ever, there is a significant difference, perfectly
matched kidneys being much less likely to be re-
jected than less ideally matched organs (9).

Case Discussion. Using the autonomy-pre-
serving principle of paternalism, should David's
choice be respected or should his donation be re-
fused? The answer is determined by the assess-

Vol. 60 No. 1

DEBATABLE DONORS' CONSENT— RHODES

ment of whether David's organ "donation" re-
flects an autonomous choice. As an autonomous
choice it must be respected, and we must accept
his donation. Viewing his decision as a nonauton-
omous act, we would have to consider how trans-
planting his kidney might affect his future auton-
omy.

Although morality usually requires us to pre-
sume that the decisions others make are autono-
mous, the choices of children are the classic ex-
ception to that rule. We always feel duty bound to
evaluate our children's acts, to teach them how to
think — a particularly Kantian example, as in his
"The Didactics of Ethics," (ref. 4) — and to protect
them from the predictable harms of their choices
(for example, maiming). We do this so that when
our children become autonomous, they will be
free of handicaps and able to do what they choose.
Adolescents fall into a gray zone where it is un-
clear whether to assume that they need paternal-
istic protection or that they are fully autonomous.
Often the justifiable presumption is that the ad-
olescent's autonomy status is somewhere between
childhood and maturity. In these cases, choices
with the most serious consequences should be ex-
amined; some of their choices should be respected
and others constrained. The autonomy-preserv-
ing principle of paternalism informs us that the
likelihood, extent, and degree to which predict-
able harms might restrict future autonomy deter-
mine whether an adolescent's choice should be re-
stricted or respected.

We are left with the same two questions: (a)
Are we entitled to assume that David's choice is
autonomous? (b) If David's choice is not autono-
mous, will donating his kidney limit his future
autonomy? David's age and the apparent unwa-
veringness of his resolve to donate might lead us
to judge that David is acting autonomously.
There are, however, several reasons for conclud-
ing that David's decision might not be autono-
mous. David's stated reason for being a donor —
that he would be a hero to his family — seems
immature. Also, adolescents are notoriously emo-
tional and easily swept away by emotional
causes. Bernstein and Simmons mention several
striking, atypical features of the adolescent do-
nor's decision-making process. According to them,
rapid decision-making and a lack of ambivalence
seem to characterize adolescents' decisions to do-
nate a kidney. In fact, 50% of the potential ado-
lescent donors hearing about the need immedi-
ately volunteered to be donors. Furthermore,
Bernstein and Simmons claim that " 'black sheep'
are particularly likely to be motivated to donate

in order to regain status in the family" (10). The
tremendous emotional component of this situa-
tion might have an overwhelming impact on Dav-
id's choice. Furthermore, David's close family,
which strongly supports his donation, might have
had a powerful coercive effect on his decision.

Being an organ donor would certainly re-
strict David's future choices. Continuing to play
football would no longer be an option for him.
Other career opportunities, in particular those
which might be most attractive to a young man
who enjoys contact sports (military service, law
enforcement), might be blocked because they typ-
ically have strict physical standards for accep-
tance. The possible psychological impact of the
kidney donation might also limit what David may
do in the future.

Determining that we should not assume Dav-
id's choice to be autonomous and that his "deci-
sion" to donate would be likely to limit his future
autonomy requires paternalistic interference to
protect David. It would prohibit transplanting his
kidney now. On the other hand, determining that
we should assume that David is autonomous
would, instead, leave us with the duty of respect-
ing his choice and require us to transplant his
organ to his brother.

Conclusion

Whatever the particular conclusion in this
case turns out to be, it cannot be generalized to
cover other cases involving adolescents, children,
infants, or fetuses. In other cases, the facts might
determine a different adjudication of the conflict-
ing principles, and in some situations, additional
principles might be pertinent. Nevertheless, in
all cases where autonomy is absent and someone
will suffer harm for the benefit of another, the
least we owe them is a serious evaluation of the
ethics of inflicting that harm and a willingness to
refuse cooperation when we recognize that coop-
erating would be immoral.

References

1. Strunk v. Strunk, 1969 [445S.W.2d 145, '46 (Ky. 1969)].

2. Hart v. Brown, 1972 [29 Conn. Supp. 368,289 A2d 386

(Conn. Super. Ct. 1972)].

3. Little V. Little [576 S.W.2d 493 (Tex. App. 1979]).

4. Kant E. The metaphysical principles of virtue, Part II of

The metaphysics of morals, trans, by James W. Elling-
ton. In: Ethical Philosophy Indianapolis: Hackett Pub-
lishing, 1983, 449-450.

5. VanDeVeer D. Autonomy respecting paternalism. Social

Theory Practice 1980; 6(2):187-207.

6. Dworkin G. "Paternalism," "Paternalism; Second

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

Thoughts," In: Sartorius R, ed. Paternalism. Minneap-
olis: University of Minnesota Press, 1983, 19-34, 105-
112.

7. Dworkin G. The theory and practice of autonomy. Cam-

bridge: Cambridge University Press, 1988.

8. Rhodes R, Burrows L, Reisman L. The Adolescent living

related donor. Hosp Ethics Committee Forum, 1992;
4(5): 314-323.

9. Fischel RJ, Payne WD, Gillingham KJ, Dunne DL,
Sutherland DER, Matis AJ, Najarian JS. Long term
outlook for renal transplant recovery with one year
function: "Doctor, what are my chances?" Transplanta-
tion 1991; (January):118-122.
10. Bernstein DM, Simmons RG. The adolescent kidney do-
nor: the right to give. Am J Psychiatry 1974; 131(12):
1338-1343.

The Physician's Experience:

Cases and Doubts

Steven M. Fruchtman, M.D., Harry Schanzer, M.D., and Myron E. Schwartz, M.D.

Cases in Bone Marrow
Transplantation
Steven Fruchtman, M.D.

Two cases in bone-marrow transplantation
are, I think, a natural continuation of the discus-
sion of donors and consent by Angelina Korsun,
Thomas Tomlinson, and Rosamond Rhodes.

Case 1: True Consent and Reliability. A 44-
year-old gentleman came to his physician with a
white count of 350,000 cells/nL. A diagnosis of
chronic myelocytic leukemia was made on the ba-
sis of bone marrow histology and cytogenetics. He
was managed medically for two years and then
referred for marrow transplantation. He had two
siblings, but only one was HLA identical and thus
appropriate for marrow donation, his 46-year-old
sister who was institutionalized with chronic
schizophrenia and was taking a multitude of psy-
chotropic medications.

Thus, one of the issues was informed consent.
Could we actually explain to the schizophrenic
sister what she was going up against in marrow
donation, and was she competent to understand
the procedure?

The other issue was reliability. After HLA
typing, further medical evaluation of this woman

Adapted from presentations at the conference "Transplanta-
tion: Ethics and Organ Procurement" at the Mount Sinai Med-
ical Center on March 13, 1992. SMF's cases are from the De-
partment of Bone Marrow Transplantation, Mount Sinai
Medical Center. Address reprint requests to the author at Box
1275, Mount Sinai Medical Center, One Gustave L. Levy
Place, New York, NY 10029. HS's cases are from the Depart-
ment of Surgery, Mount Sinai Medical Center. Address re-
print requests to the author at Box 1259, Mount Sinai Medical
Center, One Gustave L. Levy Place, New York, NY 10029.
MS's case is from the Department of Surgery, The Mount
Sinai School of Medicine. Address reprint requests to the au-
thor at Box 1259, Mount Sinai Medical Center, One Gustave
L. Levy Place, New York, NY 10029.

was required. In addition, she frequently did not
appear for her medical appointments. Sometimes
she said she just wasn't feeling up to it. At other
times, she felt that things were not appropriate at
her institution and she demanded some changes
be made before she continued to cooperate with
the procedure. Remember that once the recipient
is prepared for marrow transplantation, he or she
will no doubt die from complications of the pre-
parative regimen if the marrow donor does not
show up at the appropriate time, and if the mar-
row donation is not given on the scheduled day.

A third issue, related to reliability, was
whether or not a backup marrow should be ob-
tained from the leukemia patient in case the mar-
row donor did not show up on the designated day.
If she did not show up, the only way to rescue him
after he had received total body irradiation in
preparation for the marrow transplant was to re-
infuse his own bone marrow, with its leukemia.
The only thing we would have accomplished, if
that happened, would be to return his leukemia to
him.

So we had problems with informed consent
and with the reliability of the only marrow donor
available for this man who had leukemia.

Case 2: A Debatable Young Donor. A 4-year-
old patient with acute lymphocytic leukemia had
undergone induction chemotherapy and was put
into remission for his leukemia. In the next year,
his leukemia relapsed four times. Each time he
received further chemotherapy he went back into
remission and was referred for marrow trans-
plant to another institution.

While on the waiting list for marrow trans-
plant, the patient developed leukemia of the cen-
tral nervous system, and the marrow transplant
had to be delayed. The central nervous system
leukemia was eradicated and he was again re-
ferred for transplant. His only HLA-identical do-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

nor was his 3-year-old younger brother. The par-
ents gave consent for marrow donation, and the
patient underwent marrow transplantation. The
transplant was complicated by fungal infection
and graft-versus-host disease. Unfortunately,
three months later the patient's leukemia re-
turned. The initial marrow transplant center rec-
ommended that no further therapy be given, and
he came to this institution for a second opinion
about another marrow transplant.

We felt that perhaps — and I emphasize "per-
haps"— the patient could be helped by another
marrow transplant, because up until that point
he had only received chemotherapy. Our prepar-
ative regimen included a total-body irradiation
approach, and his leukemic clone had never been
exposed to radiation. A major issue was that the
only available marrow donor was his now 4V2-
year-old younger brother, who had already do-
nated marrow for his sibling. He would have to
once again undergo general anesthesia and mar-
row donation for the remote possibility of helping
his sibling with refractory leukemia.

For infants, marrow donation is a little more
risky than for adults. We try, because of the
young age of these patients, not to transfuse them
with blood-bank blood. However, because they
are very small, it is difficult for them to donate
blood for themselves, and thus when we harvest
their marrow, they become profoundly anemic.
They can have a hematocrit of 18%-19% post har-
vesting. This does increase the risk for the proce-
dvire.

The ethical question in this case is whether
to ask a 4V2-year-old child to once again undergo
general anesthesia and donate marrow for his sib-
ling with leukemia. The chance of the second
marrow transplant curing the patient of his leu-
kemia was small.

Child-Donor Kidneys
Harry Schanzer, M.D.

Case Presentation. A 26-year-old woman,
blood group O, who had end-stage renal disease
secondary to lupus erythematosus was treated
with dialysis beginning in March 1990. She was
placed on the cadaveric kidney transplant list in
August 1991. On February 1, 1992, a double en
bloc set of kidneys from a 10-month-old baby do-
nor who died after suffering child-abuse trauma
was offered to our institution for transplantation
to this patient. The transplant went uneventfully
and the patient initially did well. On the seventh
postoperative day she developed acute pain in the
area of the transplant, and had an increase in
creatinine. A renal scan demonstrated lack of

uptake to the medial kidney. On surgical explo-
ration, an infarcted kidney was found and ne-
phrectomy of the affected medial kidney was per-
formed. In follow-up, the patient has done well,
and is stable, with a creatinine level of 3.5 mg%
one month after transplantation.

Background. At the beginning of 1990, about
20,000 patients with end-stage renal disease were
waiting for a kidney transplant. During that
year, only about 7,500 cadaveric transplants and
2,000 living related transplants were performed.
This shortage of kidneys has become more severe
with time. In the United States the rate of kidney
transplants (number of transplants for every
1,000 patients on dialysis) has been decreasing
since 1988. This national reality is even more se-
vere in New York State.

There is, therefore, a great need for maximal
utilization of the potential donor pool.

The results of transplantation of donor kid-
neys from children younger than 4 years of age
have been inconclusive. The small size of the kid-
ney and its vessels and ureter presents problems
of adequate renal functioning mass and poses dif-
ficult technical challenges. Reports from Minne-
sota establish that long-term results of child-do-
nor en bloc kidneys are similar to those obtained
with adult-donor single kidneys. A recent report
from St. Louis showed good one-year and two-
year graft survival in five cases. The rate of tech-
nical complications nevertheless was quite high:
in three patients one kidney was infarcted, in two
of them a uninephrectomy had to be performed. In
Mount Sinai over the past eight months, we have
performed seven of these transplants. In two pa-
tients, both kidneys were lost to infarction. In two
other patients, one kidney had to be removed due
to infarction. The three remaining patients did
well, with no complications. At this time, with a
follow-up of one to nine months, five of the seven
patients are doing well without dialysis.

Ethical Questions. These cases pose a partic-
ular ethical question: During the process of in-
formed consent, should the patient be told about
the characteristics of kidneys donated by children
and the potentially more complicated course and
lower success of such transplants?

These cases also bring out a more general
question: Should patients be informed about the
quality of the organ or organs to be transplanted?

Living-Related Donation of
Liver Segments: Is It an Option?
Myron E. Schwartz, M.D.

The case of a 15-month-old boy with end-
stage liver disease probably due to congenital bil-

Vol. 60 No. 1

THE PHYSICIAN'S EXPERIENCE— FRUCHTMAN ET AL.

iary atresia is a starting point for this discussion.
Due to his condition, he has suffered developmen-
tal delay, and through the second half of 1991 he
began to develop signs of liver failure. He was
referred for transplant evaluation in December,
and was placed on the waiting list for a cadaveric
donor organ the first week of January.

The patient has been hospitalized since being
listed for transplantation, and has been deterio-
rating steadily. In order to increase his chance to
get a liver, we increased the weight range for do-
nors that we would consider, agreeing to split a
larger liver if necessary to provide an appropri-
ately sized graft. Despite this, however, we have
been unable to come up with a suitable organ, and
the patient now hovers near death in the inten-
sive-care unit. The question has been raised:
"Should we approach the family about the possi-
bility of donation by one of the parents of a por-
tion of the liver?"

Living-related organ donation has long been
practiced in kidney transplantation. A kidney
may be removed with a high degree of safety, and
with little measurable long-term effect on the do-
nor. Living-kidney donor mortality has been re-
ported in the range of 0.1% or less. Such a range
must be the case for us to justify the procedure to
ourselves; we must ever be mindful of Hippocra-
tes' admonition, primum non nocere, first do no
harm.

Safety. Whether a portion of the liver can be
removed with such a high degree of safety is a
matter of conjecture. No reported series of partial
liver resections matches these statistics, but then
liver resection has not generally been performed
on young, healthy donors with normal livers. It is
a relatively small portion of the liver, the left lat-
eral segment, that needs to be removed for this
procedure. The techniques for this procedure are
well standardized, and are similar to methods
that we use routinely in the treatment of patients
with liver cancer.

Based on our experience, we feel that the do-
nor operation could certainly be performed with
less than 1% mortality. Because of the organ's
known remarkable regenerative capacity, within
6 weeks of surgery the liver will have reattained
its preoperative size and function, so that no long-
term negative effect on the donor is anticipated.
This patient's 22-year-old mother and 28-year-old
father are aware individuals who are in good
health. In the past, they have made inquiries
about the possibility of donation to members of
our staff.

Recipient Complications. There are poten-
tial complications to the recipient inherent in the

use of a portion of a liver for transplantation. Us-
ing reduced-size grafts, that is, taking an organ
from an adult cadaver donor and cutting away a
portion of the liver to an appropriate size, has
been performed in many centers, including
Mount Sinai. Problems with bleeding and infec-
tion have arisen because of the cut raw surface of
the liver graft, but overall patient survival has
been similar to that achieved when whole organs
are used.

This drawback is also a factor in living-re-
lated transplantation. In the reduced cadaver
graft, the main blood vessels of the donor liver
can be preserved to make the connections to the
recipient vessels; in the living-related procedure,
the main vessels to the liver must of course be left
in the donor. This makes it necessary to perform
additional delicate vascular reconstruction of the
graft before implantation, slightly increasing the
risk of clotting, a serious complication that may
lead to graft failure.

Theoretical Advantages. There are theoreti-
cal advantages to living-related donation. An or-
gan from a relative will provide a better match
between the graft and the recipient, possibly re-
ducing the chance of rejection. In about 5% of ca-
daver transplants, the organs do not work, pre-
sumably because of unrecognized problems in the
donor or because of the need to preserve the liver
for extended periods of time. In living-related
donation, the donor is known to be alive and
healthy, and the graft can be taken out of the
donor and placed into the recipient almost di-
rectly, minimizing the time that the graft is with-
out blood supply. Finally, the establishment of a
living-related program would permit the elective
transplantation of candidates who are medically
stable, rather than having to wait until the recip-
ient is in extremis. In our experience, the most
significant predictor of survival following trans-
plant has been the pretransplant medical status
of the recipient.

As of today, living-related liver transplanta-
tion has been performed in only one center in the
United States. It has been undertaken in a num-
ber of other countries where there is no brain-
death legislation, thus preventing cadaveric or-
gan donation. At the University of Chicago, over
40 liver transplants have been performed. There
have been no donor deaths, although one donor,
the first, required splenectomy. Recipient sur-
vival is about 90%, a bit better than Chicago's
overall results, although favorable cases were se-
lected early in their experience. No increase in
complications has been observed in the Chicago
series.

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

Reasons for Reluctance. Other transplant
centers have been reluctant to initiate living-re-
lated liver transplantation. The primary reason
has been concern over the safety of the donor op-
eration.

The argument is also made that, with re-
duced-size cadaver grafting, there is not a great
need for the living-related procedure. Before re-
duced-size grafting, about 30% of children placed
on waiting lists for transplantation died before
they received a liver; with reduced grafts, the fig-
ure now approximates that for adult candidates,
about 10% nationally. In New York, there is a
universal donor shortage, and the waiting period
for an adult liver now averages about 4—6
months; 15% of our patients listed for transplan-
tation last year died waiting.

Concern has also been raised over the ethical
validity of consent for this procedure. As com-

pared to kidney donation, where the alternative
is continued life on dialysis, the alternative to
liver transplantation is death. Some question
whether, in this setting, donor consent can truly
be given. The coercive pressure of the situation of
greatest need, where the potential recipient is
critically ill and death is imminent if transplan-
tation is not performed promptly, would make the
voluntary quality of the consent even more ques-
tionable.

Is It an Option? We have not yet offered the
option of living-related donation to this family.
We have gone to Chicago, observed the procedure,
and feel that it is well within our technical capa-
bility. We feel the pressing need to look for solu-
tions to the critical shortage of donor organs in
New York. We welcome the opportunity to lay out
this option before the medical community in this
forum.

Cases and Doubts

Panel Discussion

Moderator: Daniel A. MoROS, M.D.

Daniel A. Moros: We'll start off with the panelists
questioning one another a bit and then bring the
audience in as well. I'll begin by asking those who
presented cases whether they have any specific
questions to address to any of the panelists on
their cases.

Steven Fruchtman: Given the benefit-risk ratio
of marrow donation for a second time in a sibling
with end-stage leukemia, how do you go about
advising the parents?

Thomas Tomlinson: As you mention in present-
ing your case, I think one of the significant issues
here is the very much lower chance for transplan-
tation success in second and further attempts. If
the marrow donation had been perhaps the first
time around, then I think I would have a problem
with it for the reasons I explained in my talk.
Given the much lower prospects of success after
the first attempt, I think presuming any kind of
consent on behalf of the child who will be the
donor is much more difficult because the proce-
dure would be more of a gamble.

I think I would also want to know whether
the donor of the second donation faces any extra
increment of risk. If so, that would also be an
important factor. If risk was simply on a par with
the first donation, then I guess I would want to
encourage an open and frank discussion with the
parents about what the actual prospects were for
success the second time around and ask them to
balance the risks to the donor.
Fruchtman: My best "guesstimate" would be that

Adapted from a panel discussion at the Seventh New York
Regional Conference on Medical Ethics on March 13, 1992.
Address reprint requests to the moderator, Associate Profes-
sor of Neurology, Mount Sinai School of Medicine, Box 1135,
One Gustave L. Levy Place, New York, NY 10029.

the risk of second marrow donation, including the
general-anesthesia risk, medically is similar to
the first donation. At that time, we had little data
to go on because there was little experience with
second-marrow transplants for refractured leuke-
mia. In first transplants for refractured leukemia,
the chance for long-term survival was 40%-50%.
We really didn't know the numbers for second
transplants. Our best "guesstimate" was given to
the parents as probably less than 10%, probably
even as low as 5%, long-term disease-free sur-
vival with the second transplant.

I can give you follow-up. The parents decided
to permit the 3-year-old to donate marrow for a
second time. The marrow was obtained. The do-
nor was listless for a few days, but eventually he
was discharged and suffered no long-term sequela
of the second marrow donation. The recipient now
is six years out from the second marrow trans-
plant with no evidence of leukemia.
Tomlinson: Nice ending.

Moros: All the cases presented have raised a
question that is not considered in the theoretical
talks: Should organ recipients be informed of the
details and risks of transplantation, and should
they be invited to accept such risks? Since both
Dr. Schanzer and Dr. Schwartz have raised this
question, I invite them to offer their opinions.
Harry Schanzer: The problem that I tried specif-
ically to attack is the problem of how much the
recipient should know about what he's getting
into and how that knowledge logistically influ-
ences the work of the transplantation community.
At first glance, one would think that the patient
has the right to know if he's going to get a kidney
of inferior quality, and he has the right to decide
if he wants it or if he doesn't want it. The problem
is that one has to assess this issue against the

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

background of the shortage of organs. Establish-
ing lists of patients who will accept certain kid-
neys or certain organs raises additional problems.
Patients who are very sick are going to take
whatever is available. Patients who are very in-
telligent are going to wait for their best chance. I
think that has some ethical implications.

I would like to hear from Dr. Schwartz
whether he tells his patients about the condition
of the liver to be transplanted into them.
Myron E. Schwartz: On the liver transplant cir-
cuit we stand behind our livers. We don't use liv-
ers that we don't expect to work. In fact, in ana-
lyzing our results and in looking at the results of
others, it's been extremely hard to come up with
solid predictors of liver function. Because of the
overall shortage of organs, certainly we've wid-
ened the criteria for acceptability of livers. We've
raised the maximum donor age, for example — the
oldest liver we've used so far came from a 71-year-
old. It worked very well. We've also broadened the
variety of circumstances of death and abnormal-
ities in either liver function tests or history of the
donor that we will accept. We found that most of
the criteria developed in the early days of trans-
plantation about donor acceptability turn out to
have no real basis in any facts when it comes to
how the livers work.

I don't know how publicly I should say this,
but if we have a 65-year-old donor I tend not to
channel that organ to a teenage recipient but
rather to an older recipient. But the livers work
regardless.

If that were not true, however, I would not be
telling patients, "You're getting a bad liver" or
"You're getting a good liver" because I think of
donated livers as a pooled resource. The patient is
listed for a transplant and they're going to get a
transplant out of this pool. Fifteen percent of pa-
tients are never going to get a transplant; they're
going to die. To me, it's not one person to one
organ; that's not the way I view it. The donated
livers are a community resource, and the patient
gets on the recipient list. We have to do the best
we can to get everybody transplanted and not be
telling people they'll be getting a good or a bad
liver.

Schanzer: We in kidney transplants also stand by
our transplants. I think the problem is that we
are always working, as somebody in this group
has said, on the cutting edge. When we decide to
widen our criteria and begin to take kidneys from
a donor who is 70 years old, we don't at the time
have the evidence that that kidney will work as
well as a younger one. So the question really re-
mains: should the patient be told about the un-

certainty of that organ compared with the norm
that we have? I think that this is where the eth-
icists would help us.

Rosamond Rhodes: We had this discussion in a
kidney transplant meeting. Somebody from the
team — I believe it was Dr. Winston — came up
with an answer that I thought was very insight-
ful. He suggested that, in terms of informed con-
sent, when patients are registered as candidates
in this program, we should inform them of the
parameters of organ acceptability; for example,
we will accept en bloc kidneys, we will accept kid-
neys from aged people or organs that may be
hours postharvest. That should be part of the in-
formed consent. But once a person accepts the
general criteria, they've given up the right to
choose an individual organ offered to them. I
thought that suggestion met both the criteria of
informed consent and the sense of justice of dis-
tribution that Dr. Schwartz mentioned.
Tomlinson: I guess I would endorse that, with the
addition that I think the patient should be able to
control somewhat what sort of organs would be
acceptable to them, especially when there are
other alternative forms of treatment available —
where going under transplant means leaving
some other form of treatment which they might
find satisfactory compared to the lower chances of
success from one kind of graft over another. So I
think patients should have some control over
that. I think it should be understood, however,
from the outset what those parameters are.
Rhodes: I don't think you can have it both ways.
A patient can have control in saying, "Well, these
are the parameters at Mount Sinai; perhaps Ne-
braska has more strict criteria because they have
a greater organ pool." But either you sign up here
or you don't sign up here. That's where the choice
comes in. You can't pick and choose between or-
gans.

Tomlinson: I don't think that that's a choice
which most patients in fact have. So that's a non-
choice. Of course, patients are not setting the pa-
rameters for the transplant team; they're setting
the parameters for what goes into their body. I
guess I think patients have a fundamental right
to choose parameters for themselves. Now,
stricter parameters may lower their chances of
actually finding a matched organ. But that
should be a risk that they are helped to under-
stand. That's part of the trade-off that they should
be willing to accept if they want to set parame-
ters.

Rhodes: If a patient were to refuse the bloc kid-
neys, would he stay on the top of the list or go to
the bottom?

Vol. 60 No. 1

DISCUSSION

Tomlinson: Well, if they refuse to accept the bloc
kidneys, they've already lowered their chances of
getting a match.

Rhodes: But do they stay at the top?
Tomlinson: Well, they haven't gotten the organ
that they require.

Rhodes: They haven't gotten the organ they re-
quire, but would they still be left at the top?
Tomlinson: Some organs that are available do not
HLA-match. It doesn't mean that I get knocked
down to the bottom of the list if they're refused on
my behalf.

Moros: Let's let the audience in a bit. I see Mar-
garet Kadian, the senior clinical coordinator of
the Mount Sinai Program for Liver Transplanta-
tion, wants to comment. Margaret?
Margaret Kadian: When patients agree to be put
on the list, we have made a contract with them
that may extend to other variables that we
haven't even addressed, for example, participa-
tion in surgery by a resident or a junior attendee,
and whether the donor had cytomegalovirus.
These are big contracts and there are a lot of vari-
ables, including the quality of the organs. Do we
tell everybody every detail? Where do you draw
the line?

Tomlinson: It seems to me that we have taken the
last generation or so to get to the point where
people are allowed to make their own choices. It
used to be that when patients came to the hospital
they were presumed to have agreed to take the
whole package. In transplantation what has hap-
pened to informed consent?

Angelina Korsun: I'd like to clarify one point. As
Margaret Kadian explained, the fact is that each
and every patient is informed of the risks. Every-
one is told, "Yes, you may have an infection; it's
likely that you'll have some sort of episode of re-
jection; the treatment may work; it may not work;
you may have graft failure; you may need another
transplant." That is, in fact, explained. Patients
are forewarned that they may need to be on ex-
tensive medication prophylactically because of
cytomegalovirus or something else.

I think one of the issues is how far do you go
in explaining the details of each and every donor?
Because you can also take it one step further and
say, "Well, I'm white. I don't want a black kidney.
I don't want a black liver." Is that right? Is that
good? Or, "I'm going to designate that my organs
will only go to a person of my choice, of my ethnic
background, of my whatever."
Fruchtman: I think that Dr. Rhodes has elo-
quently discussed the difference between auton-
omy and paternalism. I think medical dilemmas
should be shared with the patients. I find that

patients frequently help me make these very dif-
ficult decisions if I give them the information
they need. I'm not sure whether getting a "white"
kidney or a "black" kidney is a medical issue. I
haven't seen a physician who knew the donor's
race, so there's no way a patient can. Perhaps
HLA, antigens, and other medical compatibilities
may be an issue, and I think patients have a right
to that information. If a possible infection, differ-
ent from the typical things that recipients are ex-
posed to, is a medical issue, in that particular
case, patients have a right to be informed. I be-
lieve they have a right to take part in the deci-
sion-making process in such cases.
Tomlinson: I endorse that point, and I want also
to add that the patient who might want to limit
her risk by limiting the sources of organs or tis-
sues that she finds acceptable is also taking a
risk. She's taking the risk that she's going to have
less of a chance of finding a suitable organ. That's
a trade-off. It's a value-laden trade-off, and it's
precisely the kind of trade-off that informed con-
sent is supposed to protect as a right of patients to
make rather than physicians to prevent them
from making. That strongly supports as a stan-
dard that we provide information to patients and
we provide them some choices at the outset, with
full explanations of what it is they're getting into.
Schanzer: As a transplant surgeon, I try to be
practical. My interest is to transplant as many
patients who need kidneys as I can and give them
the best chance of success. Logistically, if I begin
to inform each patient of the risks they're going to
incur in getting a particular kidney, I'm going to
provoke havoc. I'm going to have to throw a lot of
kidneys into the bucket because they're going to
get old and nobody's going to take them. Or I'm
going to divide the population into patients who
will get young, good kidneys and other patients
who are not so intelligent who are going to get
terrible kidneys — which I think is unfair.

I think the patient has the right to know
what he's getting into, but not hours before the
transplant, because at that point he's emotionally
disturbed and he's not objective in analyzing his
chances. The patient has to be told at the time of
education, letting him know what he's getting
into, what we're doing, why we are doing it, what
the risks are, and what's going to happen. I don't
think it's ethical at that point to divide the pa-
tients into ones who will take good kidneys and
ones who will take bad kidneys because I think
that that's ethically unacceptable. The patient
has the right, if he doesn't want to go on our list
because we take en bloc kidneys, to go elsewhere.
But I don't think that it's practical, at the end of

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

the road, to subject the patient to the dilemma,
"Do I accept this kidney or not?"
Rhodes: I agree with almost everything Dr.
Schanzer says, but I want to make a distinction
here. Usually we let people have choice about
what they want, but we're not in the same kind of
situation of scarcity. Because there aren't enough
organs to go around, they're being distributed as
a social resource, distributed according to what
purports to be a just policy for distribution. And
in such a situation, we have to accept certain lim-
itations. If you want to be part of the system, you
can. If you refuse being part of the system, don't
get a kidney in our system. Go someplace else.
There might not be any other place. So I think it's
appropriate to say such things to these patients,
while we wouldn't say such things to a patient
who's coming in for an appendectomy.

We can inform them of what the social policy
is. And here's where I differ somewhat with Dr.
Schanzer. Perhaps even just before the trans-
plant, if a person knows that this particular organ
is one of those — whatever "those" is — of a rele-
vant medical kind, the patient can refuse to be
part of the system at that point too. Just like any-
body could decide at any point that they don't
want to take any liver at all. Even if it's the best

one, they can decide that they don't want to go
through with it. A person can decide that they
won't take this kidney. But then there are conse-
quences to be borne: they get treated like some-
body else starting fresh on the list. They may not
live long enough to get their organ, or they'll have
to live for years more on dialysis treatment. So 1
would like to tell them before they get the trans-
plant, but certainly tell them after they get the
transplant, "Well, you've gotten bloc kidneys and
you may have these extra problems which we're
going to be looking for and paying attention to."
Tomlinson: If you think that they should be told
before they have the transplant that this organ
may pose a special risk, then great. Then we've
got informed consent.

Rhodes: But they're not picking and choosing.
Tomlinson: Yes, they are.

Rhodes: They're not picking and choosing this
kidney rather than the next one.
Tomlinson: Well, of course they don't pick and
choose; they refuse. That's all they've ever beer
able to do. That's all I was proposing, that they be
able to refuse depending on what their own as-
sessment was of relative risk and benefit of con-
fining or limiting the pool of organs that they're
willing to accept.

Those Who Don't Give

Dena S. Davis, J.D., Ph.D.

Imagine that you are the transplant coordinator
for your hospital. One of the people on your list is
in dire need of, say, a new heart, but his other
organs are healthy. In the course of his preadmis-
sion studies, as part of a general discussion of
advance directives, you ask if he would be willing
to donate his healthy organs to others in need,
should his heart transplant fail. He refuses. We
can imagine that his refusal is based on one of a
variety of reasons:

• perhaps he fears that the medical staff will be
less committed to fight aggressively for his life
if they know that they can harvest his organs
for others

• perhaps he belongs to a religious group which
believes that only intact bodies will be resur-
rected

• perhaps he is being solicitous of his family
members, who have themselves declined to
sign donor cards and who are just gut-level un-
comfortable with the idea of strangers walking
around with their son's body parts

As transplant coordinator, you are painfully
aware of the acute shortage of organs; every
week, you see people die who had had a good
chance of life if an appropriate cadaver organ had
been available. How do you feel about this young
man? Perhaps, if only at the subliminal level, you
find yourself shading the various subjective ele-
ments of patient selection against him. You
might wonder aloud if someone who does not trust

Adapted from the author's presentation at the conference
"Transplantation: Ethics and Organ Procurement" at the
Mount Sinai Medical Center on March 13, 1992. Adapted by
permission of Kluwer Academic Publishers from the author's
article "Organ Transplants, Foreign Nationals, and the Free
Rider Problem," Theoretical Medicine 1992; 13(4):337-347.
From the Cleveland-Marshall College of Law. Address reprint
requests to the author at Cleveland-Marshall College of Law,
1801 Euclid Avenue, Cleveland, OH 44115.

the medical staff will really comply with his post-
transplant regimen, or question if someone whose
religion forbids organ donation will do well psy-
chologically as an organ recipient. Or do you wish
you could institute a hospital or even network-
wide policy requiring that only people willing to
give are eligible to receive? What kind of policy
would you want to formulate?

If we are uncomfortable with this hypotheti-
cal patient, it must be because our gut sense of
fairness is outraged. We perceive this patient as a
free rider: someone who wants something for
nothing, who wants the benefits the system can
provide but is not willing to "do his share."

I propose to analyze this notion of the free
rider, and to show that it is intensely problematic.
I will do this in two stages. First, I will raise some
questions about the distinctions between "don't,"
"won't," and "can't," and then I want to show how
our gut intuitions about free riders fail in the con-
text of organ allocation.

Don't, Won't, and Can't

What do we mean by someone who "won't"
donate an organ after death? Whom would we
place in this category? The easiest person to place
is perhaps the hardest to imagine: the organ re-
cipient who says, "I don't want to donate — I have
no particular reason — and I am too lazy to think
about it." Moving along the continuum, we might
be a little less quick to categorize the person who
admits that she ought to think about it, but
claims that the whole subject "gives me the shiv-
ers," so she refuses to pursue it further. If she or
a loved one were about to become a transplant
recipient, we might plausibly argue that it be-
hooves her at least to confront her reluctance to
think the issue through.

What about the person with squeamish fam-
ily members? I recommend highly a short story by
Richard Seltzer, "Whither Thou Goest," which de-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

picts a young widow who had authorized the do-
nation of her husband's organs, only to become
obsessed with the thought that her beloved's
heart was still beating away in another man's
breast. She was unable to get on with her life
until she had tracked down the recipient of her
husband's heart and persuaded him to allow her
to hear it beating one more time. Perhaps our
hypothetical patient has reason to fear that one or
more of his loved ones will react in similar fash-
ion. Is that a good reason for his refusal? Do we
count that as "won't" or "can't"? It may be helpful
to remember that when we are dealing with a
living potential kidney donor, we are quite con-
servative about the risks we will allow them to
take, rejecting anyone for whom donation pre-
sents a higher than usual risk. Again, we tend to
say, "This person can't donate," although the re-
ality is that she could donate, simply at a higher
risk.

Finally, moving toward the other end of the
won't-can't continuum, consider the person with
religious objections to giving but not to receiving
organs. Normally, at least if our speech habits are
any guide, we think of religious objections as if
they were involuntary. We are more likely to
hear, "I can't eat at that restaurant, I keep ko-
sher," than "I won't eat at that restaurant, I keep
kosher." The reasons for this are complex: for one
thing, we tend to think of religious beliefs as ir-
rational— not susceptible to proof and persua-
sion— and also as inculcated in us by birth and
upbringing. For another, we use "can't" rather
than "won't" language to signal not true lack of
volition, but the pervasive belief in our society
that a person ought not to be forced to choose
between her religious beliefs and certain basic
goods. Thus someone who quits her job because
she is a Sabbatarian and her plant has moved to a
six-day work week is categorized for the purposes
of unemployment insurance as being unable to
work, as if she had carpal tunnel syndrome or
respiratory distress.

The Problem of
Foreign Nationals:
Test Case for the
Free-Rider Argument

Somewhere between "don't" and "can't" we
find foreign nationals, those people who come to
the United States solely for the purpose of receiv-
ing an organ transplant that would not be avail-
able to them at home. Since this group has been
the focus of much public scrutiny leading to var-

ious policy statements, I focus on them as my test
case for whether the free rider argument actually
works.

The main argument against giving organs to
foreign nationals is that, since they are not mem-
bers of the community which gives organs, it
would be unfair to allow them to receive such a
scarce resource. At least in theory, every organ
given to a foreign national is one organ fewer for
Americans who are dying for lack of an adequate
supply. There is little or no reciprocity, because
Americans rarely go elsewhere in the world to
receive transplants.

A free rider is someone who wants something
for nothing, who wants to avail herself of a ben-
efit which exists only because others have paid for
it. Most systems have enough room at the margin
that an occasional free rider does not bring the
entire enterprise toppling. For example, a person
who sneaks onto a bus without paying is exploit-
ing the fact that most of the passengers have paid,
which is what funds the bus in the first place.
Absent some mitigating circumstances, we say
this person is acting unjustly. Not only is she not
bearing the same burden as the other passengers,
but over time such behavior drives the costs of the
system up, so that other passengers pay even
more.

Voluntary and Involuntary Free Riders. As-
suming our passenger is able to pay, this seems
like the classic example of unjustified free rider
behavior — pure stealing. The passenger's behav-
ior is voluntary on both sides: she chooses the
benefit of the bus ride, and she chooses not to pay
for it. But other examples become more ethically
complicated, exhibiting a mix of voluntary and
involuntary factors.

When philosophers discuss involuntary free
riders, they tend to concentrate on the case of the
person who chooses not to pay, but who is the
involuntary recipient of an unsought benefit,
which involves a distinction between "accepting"
(more active) and "receiving" benefits (1). Imag-
ine a group of homeowners who live in a square
surrounding a weed-choked, vacant lot — a para-
ble suggested by a somewhat difi'erent example in
R. Nozick (2). A few energetic yuppies in the
group convince the others that it would be worth-
while to contribute some money and elbow-grease
to clean up the lot, repair the fence, and plant
flowers. This would heighten the homeowners' es-
thetic enjoyment of their homes, as well as in-
creasing the resale value of their property. But
one curmudgeon holds out. Although he has as
much time and energy as his neighbors, he does

Vol. 60 No. 1

THE FREE-RIDER PROBLEM— DAVIS

not care about esthetics and plans never to sell
his house.

In one scenario, our curmudgeon likes flow-
ers as well as the next person, but simply chooses
to put his time and energy into other goods which
he values more. In that case, he truly does benefit
from his neighbors' sacrifice, unless he covers his
eyes and holds his nose each time he passes the
garden.

In the other scenario, he is like Oscar the
Grouch and prefers the weed-ridden lot, or at
least is indifferent to it. If his neighbors go ahead
without him, he becomes an involuntary free
rider: through no choice of his own, he is the re-
cipient of benefits for which he did not pay his
share. (Of course, his free rider status is imposed
on him only in a qualified sense. Should he find it
ethically abhorrent to be put in that position, he
could choose to contribute his share even though
he does not want the benefits — "so-called bene-
fits," from his point of view.)

The situation of foreign nationals in need of
organs is the exact reverse of the problem of the
grouchy curmudgeon. The curmudgeon chooses
not to contribute, but has the benefit thrust on
him. The foreign national wants the benefit — the
donated organ — but through no choice of her own
is unable to "pay her share." That is to say,
through no choice of hers, she lives in a country
which does not have a transplant program. This
lack may be due to scarce resources, slow techno-
logical development, or, as in Japan, a cultural
understanding of the definition of death that does
not enable the harvesting of cadaver organs. The
foreign national might well prefer not to be a free
rider, she might prefer to have been born in a
country which had its own transplant program,
and she might have supported such a program
enthusiastically, but the country of our birth is
something we do not choose.

The Concept of Group Membership. The free
rider argument should not be used to exclude for-
eign nationals from receiving organs, because, as
I will show, it is based on some unexamined as-
sumptions about group membership. The classic
free rider argument focuses on an individual who
makes a choice, for example whether to help with
the flower garden. The free rider argument in the
context of organ transplants, however, focuses on
groups rather than individuals, without making a
logical argument for how the group is defined.

Arguments against sharing organs with for-
eign nationals are based on the claim that they
are not part of the "giving community." As Prot-
tas puts it:

It is a sad fact that as long as a shortage of organs contin-
ues, some individuals in need must be denied a tran.splant.
Under these circumstances members of the giving commu-
nity (both American citizens and aliens living in the United
States) have a right to expect that their medical needs will
be met and that patient selection decisions will not be made
to their detriment (3).

Because the argument depends so strongly on
the notion of the "giving community," we need to
ask why it is defined as all American citizens and
resident aliens, rather than in some other fash-
ion. By choosing geography as the limiting crite-
rion, Prottas et al. include some and exclude oth-
ers without explaining why this is an appropriate
choice of boundaries. The answer that leaps to
mind is that citizens and residents pay taxes, ac-
cording to a formula worked out through the dem-
ocratic process; these taxes form a significant part
of the fiscal support of the research and imple-
mentation of the transplant process. But the tax
argument does not work, because it is agreed by
everyone in this debate that foreign nationals
must pay their own way.

Money, therefore, is not the criterion of ex-
clusion, because money is something the foreign
national can bring with her. The real criterion is
the organs. Americans, it is argued, have been
donating their organs in acts of generosity and
self-sacrifice. "Only the willingness of family
members to put aside their own grief can result in
the organ retrieval. Every organ transplant
starts in tragedy and kindness" (4).

But does American residence automatically
define an individual as part of the giving commu-
nity? A 1985 Gallup poll discovered that 75% of
Americans declared themselves willing to donate
their organs, but only 17% had actually com-
pleted donor cards. A poll commissioned by
UNOS in 1987 came up with different statistics:
50% were willing to donate their own organs, and
63% were willing to donate relatives' organs, but
only 23% had signed a donor card (5). In other
words, even among those Americans without re-
ligious or other objections, most people are too
indifferent, or too reluctant to confront the
thought of their deaths, to take a few minutes to
sign their names. Given the easy opportunity
(which they can revoke at will) to declare them-
selves willing, they decline. Why are they mem-
bers of Prottas' "giving community," while some-
one from Japan, with no opportunity to give, is
excluded?

Since Prottas' definition fails, what unit of
society in America could function as the logical
equivalent of the individual free rider? The indi-

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

vidual transplant candidate herself might be a
defensible choice, at least insofar as the person
had embraced or rejected the opportunity actively
to join the giving community, for example by be-
ing a blood or plasma donor, signing a donor card,
or working as a blood bank volunteer. In all the
discussion of point systems and allocation crite-
ria, I have never seen anyone make this sugges-
tion.

If we balk at making the individual the ap-
propriate unit, perhaps the family is the proper
candidate. It used to be common in hospitals for
patients to be charged for each unit of blood they
used, but those charges were "forgiven" for each
unit donated by friends or family in their name.
Paul Ramsey suggested something like this when
he proposed that "families that shared in premor-
tem giving of organs could share in freely receiv-
ing if one of them needs transplant therapy" (6).
This idea is appealing, especially if it motivates
people to donate. But I doubt that we would want
to carry it to its logical conclusion. Imagine a
young adult in need of an organ transplant, who
was brought up in a religion which objects to do-
nation and who has only recently changed her
beliefs. Her friends and family, all adherents to
the tradition, have refused to donate, perhaps
even preached against it. She has no "credits"
built up in her "giving" bank. Although there
might be some cynical snickers about "Christian
Scientists with appendicitis," would anyone seri-
ously argue that she ought to be disqualified as a
transplant candidate? On the contrary, I think
most commentators would be quick to point out
that one cannot choose one's family, and that one
ought not to be penalized for things over which
one has no control.

A more difficult version of this question is
posed if we contemplate people whose religious
beliefs allow them to receive but not to donate
organs. Should an American citizen be refused an
organ because neither he nor his coreligionists is
willing to donate? At first glance, this might ap-
pear "fair," but why should one person's religious
objection to organ donation preclude her from re-
ceiving one, when another person's lazy indiffer-
ence to signing a donor card is not taken into
account?

In fact, there is little or no logical relation
between the idea of the "giving community" and
the reality of American residency or citizenship.
Among the majority of Americans who approve of
donation in principle, there is no reason to believe
that actual recipients or their families were any
more likely to have taken active steps toward

supporting the transplant enterprise than the av-
erage American. People are assigned to places on
the waiting list, there to await their fate, without
regard to their status as "givers." The only argu-
ment in favor of identifying the giving unit with
American residency is that it is nonintrusive (a
matter of public record) and conceptually simple,
a "bright line rule." But the argument of conve-
nience is not powerful enough to be persuasive in
a life-and-death context.

Conclusion

There is a prudential argument for excluding
foreign nationals: the fear that Americans will
perceive the allocation system as unfair, and the
whole system of donation, which is based on gen-
erosity and good will, would founder. The evi-
dence for this is ambiguous at best. Where public
resentment does exist, it is often based on the
belief that foreign nationals have been given pref
erence, or have "bought" their way to the top oi
the list. There are anecdotal reports of the Amer-
ican public pulling away from donation because ol
indignation over foreign patients. However, a poll
commissioned by UNOS in 1987 discovered that
more than half the respondents thought that re-
cipients should be selected on the basis of need;
one third felt that U.S. citizens should be treated
first. Only 22% were willing to deny organs on the
basis of nationality, and that group was selective
about which countries were disfavored (7). Thus
the case for exclusion rests primarily on a notion
of group membership which categorizes the for-
eign national as a free rider. As I have shown,
this is not a defensible position.

Let us look again at the hypothetical
"don'ts," "won'ts" and "can'ts" I described at the
beginning of this essay. We saw people with reli-
gious objections, people who feared that donation
might have a deleterious effect on their loved
ones, and people who were born in countries with-
out organ programs. Can we seriously argue that
these people are less deserving of a shot at life
than the large percentage of Americans who say
that they are in favor of organ donation, but can't
be bothered — or can't face up to inchoate anxi-
eties— to sign a donor card?

References

1. Simmons JA. Moral principles and political obligations

Princeton, NJ: Princeton University Press, 1979.

2. Nozick R. Anarchy, state and Utopia. New York, NY: Ba-

sic Books, 1974.

3. Prottas JM, et al. Statement of Exception (to Chapter V ol

U.S. Dept. of Health and Human Services, Orgar

^^ol. 60 No. 1

THE FREE-RIDER PROBLEM— DAVIS

Transplantation: Issues and Recommendations (Report
of the Task Force on Organ Transplantation) April
1986).

4. Prottas JM. In organ transplants, Americans first? Hast-

ings Center Report 1986; 16:23.

5. The National Organ Procurement and Transplantation

Network. UNOS policies regarding transplantation of
foreign nationals and exportation and importation of

organs 1988. See also JM Prottas and HL Batten, The
willingness to give: the public and the supply of trans-
plantation organs. J Health Politics Policy Law 1991.

6. Ramsey P. The patient as person. New Haven, CT: Yale

University Press, 1970, 123.

7. Evans R, Manninen DL. U.S. public opinion concerning

the procurement and distribution of donor organs.
Transplantation Proc 1982; 20.

Legal Aspects of the Sale of Organs

Beth Essig, J.D.

With the introduction of transplants the de-
mand for organs has grown and become more
compelling. Health care professionals have strug-
gled to find ways to increase the supply of trans-
plantable organs. One of the more frequently dis-
cussed options is purchasing them; that is, paying
donors for organs.

The legal aspect of the sale of organs is
straightforward. Donors of human organs may
not be compensated for the donation. In 1984, the
National Organ Transplant Act was enacted by
Congress (1). This federal statute flatly and
clearly prohibits the purchase or sale of any hu-
man organ (2). In 1988, the statute was amended
to prohibit the sale of fetal organs.

The act prohibits a transfer of a human "or-
gan" in interstate commerce for "valuable consid-
eration." It defines human organs as "kidney,
heart, lung, pancreas, bone marrow, cornea, eye,
bone, skin or any subpart thereof or any human
organ" (3). The statute does permit "reasonable
payments associated with" obtaining and process-
ing the organ, including "travel, housing and lost
wages incurred by the donor" (4). The statute does
not prohibit compensating a blood donor. Most
states have parallel statutes.

History of Prohibition of Sale
of Organs

Restrictions on the trade in human organs
and body parts are premised on the fundamental
concept that people (and their next of kin) retain

Adapted from the author's presentation at the conference
"Transplantation: Ethics and Organ Procurement" at the
Mount Sinai Medical Center on March 13, 1992. From the
Office of the General Counsel, Mount Sinai Medical Center.
Address reprint requests to the author, Beth Essig, Esq., Vice
President and Associate General Counsel, Box 1099, Mount
Sinai Medical Center, One Gustave L. Levy Place, New York,
NY 10029.

a property interest in their body after their death
Prior to the early 19th century, there was no rec-
ognized property interest in the human body, and
body parts and human bodies were traded with
relative impunity. In fact, the trade in body parts
was so brisk in the 19th century that "resurrec-
tionists" roamed the streets harvesting corpses
and selling them to schools of surgery. A murdei
committed by several overzealous resurrection-
ists to increase their inventory resulted in tht
British Parliament's enactment of the first com-
prehensive legislation governing the market in
human organs. The law set out two fundamental
principles: (a) that bodies and body parts coulc
only be received by legitimate licensed schools for
the purpose of training future surgeons; and (b
that it is the family or next of kin which governs
the distribution of the body. Modern statutes thai
address the sale of human body parts still do so or
the basis of this framework laid out in the 19tl:
century. The most important aspect of the law is
that it recognizes that there is a property interest
inherent in a dead body. The recognition of the
body as property that can be devised or donatec
by the individual or controlled by his family is
still the law today.

In 1968 the Uniform Anatomical Gift Act
(UAGA) was approved by the National Confer-
ence of Commissioners on Uniform State Laws
an advisory body that develops model legislation
which it recommends to the various state legisla-
tures for enactment by the states. Within five
years of its approval, virtually every state had
enacted a substantially similar statute (5). With
the widespread adoption of state statutes mod-
elled on the UAGA, and the federal government's
enactment of the National Organ Transplant Act,
the current system of donation of organs became
firmly entrenched in statutory law. These laws
contain essentially the same basic elements as
the law that had developed in 19th century En-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 199S

Vol. 60 No. 1

SALE OF ORGANS— ESSIG

gland. That is, the law (a) restricts who may re-
ceive donated organs and (b) recognizes that the
donor (and his next of kin) can control whether
and how the body parts can be donated.

It is somewhat anomalous that while recog-
nizing a property interest in the human body, the
"owner" may not receive compensation for its
sale. One of the few recent court cases to consider
the question of the property interest of a person in
his body is Moore v. Regents of the University of
California (6). John Moore, a patient at UCLA,
had had his spleen removed. He subsequently
learned that tissue from his spleen had been used
to develop a cell line of some commercial value.
Mr. Moore, considering himself the owner of his
body, asserted that he was entitled to financial
compensation. Indeed, the trial court determined
that Mr. Moore had a "property interest" in the
cell line and ordered that he be compensated. This
decision was reversed by the California Supreme
Court, which determined that, assuming that ap-
propriate consent had been obtained from Mr.
Moore for the use of his spleen for research pur-
poses, he did not retain ownership interest after
the removal of the organ and was not entitled to
compensation.

The Current Debate

One of the most significant issues for profes-
sionals involved in transplants is how to increase
the supply of organs. In that context, numerous
proposals have been made. These options are dis-
cussed in other articles in this symposium. They
include offering compensation to the estate of the
donor, or to the living donor, for the purchase of
the organs.

Proponents of permitting the sale of human
organs argue that a free market in organs would
increase the supply. They further argue that, in-
asmuch as the body is recognized "property" that
is controlled by the individual, it is paternalistic
to prevent the body's "owner" from seeking com-
pensation in the event of an organ donation.

On the other hand, permitting commerce in
human organs raises numerous troubling issues.
The question in this era of transplants is, can we
afford to ignore any solution that might lead to an

increase in organ availability and an increase in
the number of lives saved through transplants?

There is a concern that the availability of
compensation for organ donors would exploit the
poorest and most vulnerable members of society
and pressure them inappropriately to contribute
their organs. This view holds that compensation
interferes with "voluntariness," and that organ
donors, especially live donors, who agree to do-
nate their organs for money will be unduly influ-
enced by the availability of compensation. This
concern is visible as well in other areas of medi-
cine where compensation for nontherapeutic
health care or anything that interferes with an
individual's bodily integrity is restricted or pro-
hibited, for example, in the well-accepted princi-
ple that a payment to induce a research subject's
participation in a protocol is inconsistent with the
voluntariness of the informed consent process for
human subjects.

The view that payment for the donation of
the body interferes with "voluntariness" also un-
derlies the recent spate of legislation and case law
which prohibits the payment of money to "surro-
gate mothers" or which makes agreement for pay-
ments to surrogate mothers unenforceable. The
position taken by some in that debate is that the
payment is unduly coercive and will induce
women to make decisions to become surrogate
mothers solely for economic advantage.

Conclusion

The view that donors should not be compen-
sated is deeply ingrained in our legal tradition.
Nonetheless, the need for organs to save lives
may lead to a reexamination of that view and a
relaxation of the current prohibition under appro-
priately regulated circumstances.

References

1. Pub. L. No. 98-507, 42 U.S.C. §§ 273-274(e) (1991).

2. 42 U.S.C. § 274(e).

3. 42 U.S.C. 274(c).

4. 42 U.S.C. §274e(c)(2).

5. See Uniform Anatomical Gift Act published in Uniform

Laws Annotated (1987).

6. 51 Cal. 3d 120, 793 P. 2d 479 (1990).

Markets and Morals:

The Case for Organ Sales

Gerald Dworkin, Ph.D.

Arthur Caplan has said that "perhaps the most
pressing policy issue facing those within and out-
side of the field [of organ transplantation] con-
cerns the shortage of organs available for trans-
plantation to those with end-stage organ failure"
(1). The options available to increase the supply of
scarce goods are basically three — donation, con-
scription, or sale. A good deal of attention has
been focused on the first two methods (I take pre-
sumed consent to be basically conscription with
an option to opt out before death), but the sale of
organs has been little discussed.

I focus on the issue of whether there are good
arguments of an ethical nature which rule out a
market in organs. I leave to one side discussion of
whether such markets would in fact increase the
supply of organs, whether there are practical dif-
ficulties in the implementation of such a scheme,
whether political considerations (in the broad
sense) would make it difficult to gain support for
such a system. My only task today is to assess the
moral arguments.

The first distinction we must make is be-
tween a futures market and a current market —
that is, between the decision of an individual to
sell the right to his organs after his death, and the
decision to sell organs while he is alive. I assume,
for the sake of this discussion, that if there are
moral objections to the sale of organs they will
take their strongest form against the sale of or-
gans from living donors. Hence if one can show
that there are no conclusive arguments against
such sales, one will have shown, ipso facto, that

Adapted from the author's presentation at the conference
"Transplantation: Ethics and Organ Procurement," at the
Mount Sinai Medical Center on March 13, 1992. From the
Department of Philosophy, University of Illinois at Chicago.
Address reprint requests to the author at 712 Judson Avenue,
Evanston, IL 60202.

there are no conclusive objections against the sale
of cadaver organs.

I first briefly consider the arguments in favor
of a market in organs and claim that in the ab-
sence of moral objections, there is no reason for
not having such markets. I then want to consider
all the plausible arguments against the sale oi
organs and show that they are not legitimate ob-
jections. My conclusion will be that, in the ab-
sence of further arguments which survive critical
scrutiny, there are good reasons for favoring a
market in organs.

Arguments for a Market

We currently accept the legitimacy of non-
commercial solid-organ donations. We also accept
the legitimacy of the sale of blood, semen, ova,
hair, and tissue. By doing so we accept the idea
that individuals have the right to dispose of their
organs and other bodily parts if they so choose. By
recognizing such a right we respect the bodily au-
tonomy of individuals, that is their capacity to
make choices about how their body is to be
treated by others. By recognizing such a right we
also produce good consequences for others, that is,
save lives, allow infertile couples to have chil-
dren, further medical research, and so on. But the
primary good achieved by such a right is the rec-
ognition of the individual as sovereign over his
own body. A market transaction is one species of
the larger class of voluntary transactions. Allow-
ing people to sell things is one way of recognizing
their sphere of control.

Finally, by allowing individuals to either
barter or sell something, we increase their level ot
well-being. Since such transactions are volun-
tary, they are presumably only engaged in when
the individual believes himself or herself better
off without the good and with the cash (or an al-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

Vol. 60 No. 1

THE CASE FOR ORGAN SALES— DWORKIN

ternative good in the case of barter) than without
the cash and with the good.

So markets can increase both autonomy and
well-being.

Arguments Against a Market

There are often compelling reasons why we
should not allow individuals to sell what they
could give. We do not allow markets in votes, in
babies, in judicial decisions, in college grades. In
these cases we recognize countervailing consider-
ations which are sufficient to overrule the consid-
erations in favor of markets. So the question
before us is whether there are such counter-
arguments in the case of markets for human or-
gans. I propose to consider the arguments that
have been adduced and show that they are not
compelling.

Exploitation of the Poor. One of the most
powerful arguments against a market in organs
is the element of exploitation of the poor. Clearly,
those who are most likely to wish to sell their
organs are those whose financial situation is most
desperate. Those who have alternative sources of
income are not likely to choose an option which
entails some health risk, some disfigurement,
some pain and discomfort. The risks of such sales
will certainly fall disproportionately by income
class.

But what exactly is supposed to follow from
these facts? Is it that, because of this, the choices
of the poor are not, in fact, fully voluntary? This
seems to me false. Or if it is true, it has a much
wider implication than that organs should not be
sold. It suggests that poor people should not be
allowed to enter the army, to engage in hazardous
occupations such as high-steel construction, to be-
come paid subjects for medical experimentation.
There are certainly objections of justice to the cur-
rent highly unequal income distribution. But it
seems to me paternalistic in the extreme, given
that injustice, to deny poor people choices which
they perceive as increasing their well-being.

Here it is important to have some idea of the
size of the risk we are talking about. One study
has estimated that the increased risk of death to
a 35-year-old from giving up one kidney is
roughly the same as that associated with driving
a car to work 16 miles a day (2). Imagine saying to
a poor person either that her choice to commute
such a distance is not voluntary, or that if it is,
she still ought not to be allowed to commute such
a distance, although we will allow middle-class
persons to do so.

To make this point more vivid, what would
your reaction be to the following proposal made

by one author in response to this objection? Pro-
hibit purchases from individuals whose average
income is less than 80% of median family income.
This has the effect of removing persons in the
lower 40% of the income distribution from the
market (3). Would you now be more, or less, in-
clined to favor organ sales?

Note also in the context of arguments about
justice that the poor are disproportionately repre-
sented among those who need transplants. Thus,
assuming — as is currently the case — that the gov-
ernment subsidizes most organ transplants, they
stand to gain as a class more than the rich.

Another objection based on the fact of income
inequality is that because of unequal bargaining
power the price paid to the poor will not be a fair
one. They will not get the full market value of
their organs. If there were evidence that this was
true, the solution would be to regulate the mar-
ket, not forbid the sale. One could establish min-
imum prices analogous to minimum wage laws.

Distributional Consequences. If organs are
for sale to the highest bidder, the rich will get
them and the poor will not.

First, this seems an objection not to the sale
of organs but to the general system of medical
care based on ability to pay. There are currently
at least 50 different types of artificial body parts
which are distributed according to ability to pay.
Why is it better for the rich to have better access
to artificial than to human kidneys?

Second, currently, few individuals pay for
transplants out of their own funds. Most trans-
plants are paid for by public and private insur-
ance. So the issue again is access to health insur-
ance, not access to organs.

Note also that the main costs associated with
transplants are likely to remain the fees of doc-
tors and hospitals and the costs of drugs, all de-
termined by markets. Why is it legitimate for
these to be the results of markets and not the
organs themselves?

But if one finds that the distributional impli-
cations are unsatisfactory, regulations or restric-
tion on sales are called for. We could adopt a
scheme, for example, in which it would be illegal
for private individuals to sell organs to other pri-
vate individuals. They could only sell them to the
state. The state then could adopt whatever
scheme of distribution would ensure justice in
transfer — perhaps a lottery among the equally
medically needy, or a first-come, first-served prin-
ciple.

Irreversibility. One objection to the sale of
organs, as opposed to renewable tissues such as
blood or semen, is that the decision is irreversible.

THE MOUNT SINAI JOURNAL OF MEDICINE

Januarj' 1993

Individuals may come to regret the fact that they
have sold a kidney — particularly if they develop
kidney problems with the remaining organ. But
we currently allow individuals to make many per-
manent changes in their body, including breast
diminishment and sterilization. If we feel the
problem is more severe we can establish waiting
periods, counselling, and so forth.

More Choices Not Always Better. The argu-
ment that more choices are not always better says
that allowing new options does not leave the old
options unaltered. Applied to the sale of organs,
the claim is that once a market price is estab-
lished for organs, individuals who choose not to
sell do so in the knowledge that they have made a
choice which leaves their family worse off eco-
nomically than they might have been. Individu-
als are choosing to decline an option which they
formerly did not have. They may be psychically
worse off than if they never had such a choice. I
agree that this is a cost. I do not see, however,
that it is anywhere near the psychic costs that are
incurred by individuals and their families who
face blindness and death as a result of an inade-
quate supply of organs.

Another psychic cost is more significant, as
Hansman argues (3). If one assumes that because
of tissue matches, the most efficient donations are
from family members, it is likely that introducing
markets is liable to strain family relations. Fam-
ily members are likely to be resentful of being
asked to contribute without compensation when a
stranger would receive substantial payment. It
seems, however, that the rapid development of
immunosuppressive drugs may considerably
weaken the first premise of this argument.

Commodifieation. Finally we come to a large
class of arguments which object to the commodi-
fieation of organs. These arguments are rather
diverse in character — many are discussed by Ra-
din (4) — and one has to examine them carefully to
see how they differ and whether any of them have
sufficient force to overcome the presumption in
favor of allowing sales of organs.

Altered Nature of the Transaction. Peter
Singer, in a well-known argument against the
sale of blood which would carry over to the sale of
organs, claims that the nature of giving changes
when blood is allowed to be sold as well as do-
nated:

If blood is a commodity with a price, to give blood means
merely to save someone money. Blood has a cash value of a
certain number of dollars, and the importance of the gift
will vary with the wealth of the recipient. If blood cannot be
bought, however, the gift's value depends upon the need of
the recipient (5).

There are actually two arguments here. The
first is that the sale of blood means that the sig-
nificance of the transfer will vary with the wealth
rather than the need of the recipient. Unfortu-
nately this argument is much too powerful, since
it is an argument against the sale of anything.
Why distinguish blood from food?

The second argument has more weight. It is
that if one adds to the existing practice of dona-
tion the use of a market, the situation for donors
is altered. Whereas before they were able to give
something that could not also be purchased, now
they can only give something that has a price as
well. The nature of their gift is changed. Al-
though I concede that this is true, I do not see it as
a compelling objection to allowing such sales. Do-
nors do not have the right to have their gift retain
its special character, and if the price of so doing is
that potential recipients of life-saving resources
are excluded from receiving them (because the
supply of donations is limited), the consequences
alone would argue for not forbidding such sales.

Alienation. Charles Fried argues that:

When a man sells his body he does not sell what is his, he
sells himself. What is disturbing, therefore, about selling
human tissue is that the seller treats his body as a foreign
object . . . the shame of selling one's body is just that one
splits apart an entity one knows should not be so split (6).

Notice first that this argument (similar to
one given by Kant) applies to the sale of blood and
semen as well as organs. So if this argument is
good, it shows that our current policies are ille-
gitimate. (Although Fried seems to take it back in
a footnote [6, p. 143] saying that the selling of
blood is "personally bad . . . though not in any
sense wrong.")

But the main objection to this argument is
that it implies not only that the sale of blood or
hair is bad, but also that the donation of such
bodily parts is bad as well. For if selling organs
splits apart an entity one knows should not be so
split, so does donating it. One treats one's body
just as much as a foreign object if one gives away
a kidney as if one sells it.

The danger we want to avoid at almost all
cost is that people start to be treated as property
by others. But this is avoided by leaving all deci-
sions about their organs, tissues, and so on to the
persons themselves, and insuring that their deci-
sions are voluntary.

Driving Out Altruism. The argument about
driving out altruism is that allowing a market in
some item will make it less likely that those who
were inclined to give on altruistic grounds will
continue to do so. The data on blood are ambigu-

Vol. 60 No. 1

THE CASE FOR ORGAN SALES— DWORKIN

)us on this point — some tending to show such an
jffect, some not. It is clear, however, that the
presence of markets does not generally drive out
altruistic motives. Most hospital workers are
paid, but there are still volunteer workers. There
are markets for used clothing, but many people
give their used clothing to the needy. Lawyers are
paid for their services, but many contribute a por-
tion of their time pro bono. Finally, even if it were
true that a market in organs would somewhat
reduce the number of people who donate organs, if
the total supply is increased, one has to weigh the
loss of altruism against the gain in human lives.
I see no reason to suppose that the balance will be
negative. After all, we allow a commercial mar-
ket for caregivers for our elderly parents — surely
an arena in which not only generalized altruism
but debts of gratitude play an important role.

Conclusions

It seems to me that if we take into account all
the welfare losses that will accrue because of the
introduction of markets for organs, it will still be
the case that if the supply of such organs is sig-
nificantly increased, the two major gains in wel-
fare (improved health and decreased mortality,
and increased income for sellers) will signifi-
cantly outweigh the losses. If there are no non-

consequentialist considerations (such as denials
of rights or considerations of justice) which might
trump such consequentialist considerations, the
consequences ought to be determining.

My conclusion is that, absent other and
stronger arguments than those considered, given
that both rights and welfare argue in favor of a
market for living organ donations, there is no rea-
son not to allow them. In addition, whatever the
force of these objections, most of them are consid-
erably weaker when applied to the sale of future
rights in cadaver organs. So such a scheme is, I
believe, certainly warranted.

References

1. Caplan A. Beg, borrow, or steal: the ethics of solid organ

procurement. In: Mathieu D, ed. Organ substitution
technology. Boulder: Westview Press, 1989, 60.

2. Hamburger J, Crosnier J. Moral and ethical problems in

transplantation. In: Johnson D, ed. Blood policy: issues
and alternatives. Washington: American Enterprise In-
stitute, 1968.

3. Hansmann H. The economics and ethics of markets for

human organs. J Health Politics Policy Law 1989;
14(lt:74.

4. Radin M. Market-inalienability. Harvard Law Rev 1987;

100.

5. Singer P. Altruism and commerce: a defense of titmuss

against arrow. Phil Publ Affairs 1973; 2:314.

6. Fried C. Right and wrong. Cambridge: Harvard Univer-

sity Press, 1978, 142.

Getting and Giving:

Panel Discussion

Moderator: Rosamond Rhodes, Ph.D.

Lewis Burrows: I would like to review some re-
marks I made at a recent meeting, "Justice, Com-
merce, and Ethics in Transplantation," sponsored
by the Canadian government and the American
and Canadian transplant societies. At that meet-
ing members of the clergy took a strong stand
against both selling organs and versions of "re-
warded giving," apparently because of belief that
human beings are intrinsically good and there-
fore, given the opportunity to be altruistic, they
will be. Unfortunately, as we have learned at this
conference, there is an increasing need for ca-
daver organs which is not being met. Someone
has commented here that the only group asked to
be altruistic is the recipient's family. Certainly
hospitals profit greatly from transplantation.
Physicians and transplant professionals who earn
their living by or advance their careers through
transplantation profit greatly. The medical-in-
dustrial complex and the drug companies make
huge amounts of money through transplantation.
Even the government profits when patients no
longer need dialysis. I cannot see anything wrong
with the concept of rewarded giving, at least as an
interim measure until we can improve our ability
to attract donors, provided that the distribution
be either through government or paragovernmen-
tal agencies to insure that no commercialization
is involved in the distribution.
Robert M. Arnold: It seems to me that those of us
who deal with organ donations and procurement
need to stop going from one failed policy to an-

Adapted from a panel discussion at the conference "Trans-
plantation: Ethics and Organ Procurement" at the Mount
Sinai Medical Center on March 13, 1992. Address reprint re-
quests to the moderator, Director of Bioethics Education,
Mount Sinai School of Medicine, Box 1193, One Gustave L.
Levy Place, New York, NY 10029.

other and spend more time trying to figure out
why the policies have failed. In the early 1980s
the claim was that people wanted to give and the
problem was that doctors didn't ask often enough.
If we would only ask, more people would give.
However, with increased requests we didn't get
the increase in donations that everybody thought
we were going to have. Now the movement is to
either pay people for giving or create a futures
market. I am of the mind that there aren't non-
consequential arguments for not having a mar-
ket. But I am not sure that throwing money at the
problem is going to solve it. We ought to begin to
do some small-scale studies to figure out why peo-
ple give, and what things will increase giving. It
may be that giving would increase if only people
from organ procurement organizations asked.
Bernard Baumrin: I'll throw my two cents in. I
think a cheaper, easier, and more effective solu-
tion would be, if you don't give you don't get.
People who don't sign up can't get organ trans-
plants.

Charles Miller: That's a possibility. You would
join the donor community in order to be a recipi-
ent in the same community. But the logistics
would be difficult, and the education is so fraught
with error that it may not be practical.
Burrows: Only a very small percentage of people
ever need a transplant. There are only 25,000
people in the United States today who need a
transplant. You may extend the indications in the
year 2,000, but as of now the vast majority of
people just would take the chance of not signing a
donor card because they'll probably never need a
transplant.

Miller: We had 4,000 organ donors last year and
the year before; 25,000 patients are on the recip-
ient list now. The average rate of consent in the
United States has been around 50%.

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

Vol. 60 No. 1

DISCUSSION

Arnold: The refusal rate varies. In an Ohio De-
partment of Health study the refusal rate was
70%. In a New Jersey study the refusal rate was
60%.

Miller: Sales or financial incentives might make
the donation rate go down. It could create a back-
lash against altruism. And so I agree that a pilot
study of these ideas should be done to see exactly
what does happen to the consent rate.
Arnold: In 22 hospitals in Pittsburgh and Minne-
sota we are reviewing every patient who dies in
the hospital who is eligible to donate organs, tis-
sues, or corneas. We're interviewing all health
care professionals associated with those patients'
deaths and asking if they identified the patients
as eligible, whether they talked to the families,
what the professional said, and what the family
said. A year into the study, a rough look at the
data indicates that, in fact, health care profes-
sionals do a good job of identifying organ and tis-
sue donors. The major reason for failure in pro-
curement is that families say no. In our planned
followup study we are going to interview families
to try to find out why families are saying no. With
that information you can try to set up interven-
tion.

Dena S. Davis: I think a lot of people don't sign
donor cards because they're afraid that doctors
will let them die too soon. This is ironic because
people also fear that doctors won't let them die
soon enough. People don't trust that the medical
community is communicating with them.
Beth Essig: Every year more and more countries
seem to be moving toward a policy of presumed
consent.

Arnold: But no one's done a good empirical study
on whether doctors and the organ procurement
organizations really practice presumed consent.
Arthur Caplan tells stories about France where
they have presumed consent but people don't fol-
low it. People still ask families, and when fami-
lies say no they don't take the organs. In Austria
this doesn't occur, but Austria has a long tradi-
tion of using dead bodies for state purposes. In
fact, one of the things that's interesting about
Austria is that when they passed the presumed
consent law there was no opposition.
Burrows: In the United States last year there
were roughly 4,000 donations, and the average
consent rate in most states was 50%. That means
we were turned down by about 4,000 families. But
we've estimated that somewhere between 14,000
and 20,000 individuals are already awaiting or-
gan donations. This leads us to the issue of the
paid living unrelated donor. Nobody seems to
want to bring that up because the inherent class-

ism is abhorrent to us. But we should remember
that it wasn't the rich kids who fought in the bat-
tle fields of Vietnam, it was the poor kids. The
risks there were greater than the risks of giving a
kidney. So the precedent has been set to some
degree. We have been willing to tell a certain .seg-
ment of our population, "You can go out and fight
for us and take great risk with your lives in Viet-
nam." But to date we're unwilling to take that
same segment of the population, that desperately
poor population, and say to them, "If you want to
give up one kidney where the risk is very small
and get paid for doing it, you can do it. It's against
the law to do it." We did it again in Iraq. We said
it was unfair in Vietnam and yet who did we send
to Iraq? We didn't conscript college kids.
Miller: I have been thinking about some of the
comments Prof. Dworkin made. He was wonder-
ing why it was any different to sell an organ than
it might be to do any service or sell anything else.
There are actually some qualitative differences.
First, selling a body part is somewhat extraordi-
nary. And it may not be possible to really give
somebody information about what the risk is.
Maybe for kidneys it is, but probably for the liver
it is not. Second, it's not something you can do
over and over again. You can only do it once. I am
not sure that rules it out, but it does make for a
difference. Do you have a right to sell a vital body
part?

Dworkin: I am not sure that we are not able to get
reliable estimates of the risks. If that's true, why
doesn't that count as much against allowing peo-
ple to donate as allowing them to sell? Presum-
ably you want to obtain informed consent in both
cases. Presumably in both cases you want the de-
cision to be as fully voluntary as it can be. How
does the addition of monetary compensation af-
fect the question of informed consent?
Miller: It adds a bit of financial coercion, and
where there's a degree of uncertainty about how
reliably you can inform somebody, any addition of
coercion becomes all the more problematic.
Dworkin: But then, presumably, something of the
same worry arises, for example, with siblings and
the emotional coercion that goes with family.
Miller: Absolutely.

Burrows: I want to tell a true story which may
illustrate some of the things I have been trying to
say. About ten or twelve years ago I was ap-
proached by a well-known Indian nephrologist
who had trained here at Mount Sinai and who
then ran a large dialysis program in Bombay. In
India, of course, only the wealthy are dialyzed.
It's not universal, as in this country. He was of-
fering me the opportunity to transplant wealthy

THE MOUNT SINAI JOURNAL OF MEDICINE

January 199:

Indians who would come to Mount Sinai along
with their donors, who were being paid a consid-
erable amount of money in Indian terms. He was
hoping to use a major American center for the
best of medical care, to provide the donors with
the smallest risk and the recipients with the best
opportunity for success. All of the donor's medical
care would be taken care of by the recipient. And
I turned him down cold.

I turned him down primarily because it was a
matter of using the desperately poor in what
amounted to a bidding war by the rich. The
method of distribution was not fair. The Indian
nephrologist got upset and asked if I had any idea
what that money would offer to the poor Indian
peasant; did I realize how it would change not
only his life but the lives of his children? The
seller would be able to have a dowry for his
daughters, he would be able to buy a piece of land
for his sons, he would be able to send his children
to the university. We still turned him down, and
I don't think Mount Sinai would have accepted
my participation either. He was using the desper-
ately poor to serve the desperately sick rich.
Question: I know we won our victory over the
Soviet Union, but that doesn't imply that every-
thing is marketable. You argued that maybe you
can sell part of yourself but, would that be mor-
ally right?

Dworkin: To make my position clear: I never ar-
gued that all things should be for sale. In fact, I
gave several examples of things that I thought
should not be for sale: votes, judicial decisions,
and grades. In each case I should hope I would be
able to provide you with some kind of principled
reason as to why I wanted to distinguish those
cases from other cases. Now it seems to me that
the case you give is a perfectly reasonable one. I
don't think we should allow people to simply ex-
ercise arbitrary power over one another and force
them to degrade themselves and sell their hair,
for example. Some of you may know the play by
Durenmatt, The Visit, in which a woman comes
back to town who early in life was badly treated
by the villagers in her town. She wants to see how
much it takes to make each of these people do
some terrible thing. The question simply was how
much of a price each individual would exact. I
would certainly be opposed to allowing people to
offer large sums of money so that people could
mutilate themselves arbitrarily, but this case
seems quite different — no important social good
was served. I am assuming that organ donation
will, in fact, save people from dying, allow people
to see who otherwise wouldn't see, and so forth,
and so donation offers some important social

good. It seems to me that once this has been es
tablished, and if you accept some kind of an ini
tial presumption that people should have a righ
to control their bodies, you require a strong argu
ment to prevent the sale of a body part.
Question: What about looking at whether the re
cipient has certain rights to a transplant? Do ou:
social policies infringe on those rights?
Dworkin: At most what could be shown is tha
you do have a right to a certain institutionalize!
system of medical care, and that may include i
right to be placed on a list under certain condi
tions. But you certainly don't have an individual
ized right against the other particular person
That's clear. And unless you believe that somi
system of conscription is legitimate, I don't thinl
you even have a generalized right that somebod;
or other provide you with an organ.
Rhodes: We usually think people have at least i
right to try to get what they need. They may no
have a right to what they need, but they at leas
have a right to try to get it. But here the govern
ment says no, you can't try to get the transplan
which you need.

Essig: If society decides that procuring organs fo
people who need transplants is more importan
than letting people be buried with all of thei
body parts, things would work differently. A lot o
the things we have talked about are alternatives
We have talked about cadaver organs, but th
dead do not have the same right as the living
What we are talking about is the right to tak
organs from dead people.

Arnold: I am not sure how this right would apply
It would be like people saying, "There isn'
enough food around and yet the farmers aren'
using every inch of the farmland to grow food
Since I have a right to get what I need and I nee(
food, I should be able to force the farmer to use thi
rest of the farmland to grow food." I don't under
stand.

Rhodes: She's saying we all have a right to try t
get food. If we were allowed to buy food, thi
farmer would have an incentive for growin]
more.

Baumrin: If someone was withholding land fron
tillage for some reason and there were actuall;
people starving, the government would be enti
tied to interfere on their behalf in expropriatinj
land from the individual who kept the tillabl
land out of production.

Arnold: The government might be able to do that
although there might be countervailing reasons
What the government needs to say in this case i
look, you have a right to try to get what you want
If there were no countervailing reasons not to al

Vol. 60 No. 1

DISCUSSION

low people to sell organs, we would let them sell
organs. But, in fact, we think there are counter-
vailing reasons. Then the question is, are the
countervailing reasons consistent with societal
policy? And then we're back to the fireplug.
Essig: The first time Congress made clear that
!you couldn't sell organs in interstate commerce
was in 1986. The only state I know of that now
has a statute permitting sales is Mississippi; pre-
viously there were more states with statutes per-
mitting organ sales. There are states with no stat-
utes where there have been sales, such as
California. Currently, the movement everywhere
is to prohibit sales.

Question: The distinguishing characteristic of
what you can rightly sell is rightful ownership.
For instance, you shouldn't be able to sell chil-
dren, because you don't own children. You
shouldn't be able to sell grades, because a grade is
something the professor controls, but doesn't own.
If you rightly own it, you ought to able to rightly
sell it. And I can't see a better case for ownership
than for a body part. But there's another factor
besides superstition that is operating against al-
lowing markets in organs. Technologies tend to
inherit the politics of the era that they're born
into. Over the past century in the developed
I world, there has been a general trend toward in-
creasing authoritarianism and greater antipathy
towards freedom of individuals and markets.
Later technologies have come under tighter re-
strictions.

Dworkin: I am not inclined to accept that princi-
ple that whatever you legitimately own you may
legitimately sell. In particular, for example, I
have grave doubts about whether you should be
allowed to sell your heart. And I think you own
your heart as much as you own your fingernails.
So I think there are other considerations that en-
ter; that principle is probably too strong.
Burrows: The donor should have his own advo-
cate. As the transplant surgeon, I can't be the
advocate of both donor and recipient because I
have a conflict of interest there. The donor should
have somebody to advise him, to help him. In the
case of the under-age donor that Dr. Rhodes dis-
cussed, we tried very hard to find an advocate for
the donor other than ourselves or the pediatri-
cians who were caring for the recipient. The court
ultimately became his advocate. We had to make
him a ward of the court. Obviously there is a con-
flict under any circumstance, even in cadaveric
transplantation. We no longer care for the cadav-
eric donor in his agonal phase. That is now done
by individuals who are not in any way advocates
of the recipient.

Question: Can any surgeon who is to economi-
cally benefit from performing the surgery ethi-
cally remove an organ from a living donor for
transplant into another?

Miller: We feel justified in doing living-related
transplants because they involve an altruistic gift
to a close loved one. This is a donation you can
only give once. And, as surgeons, we're pretty
well assured that it is given at the best time and
for the best reason. I am assured that the recipi-
ent will derive a benefit from the donor's risk. In
a free commercial market, the donation might be
more valuable next week. The price of the organ
may go up. Possibly the donor should wait a week
to get the greatest benefit from donation. I could
not be certain that he was receiving his fair share
of the deal, so I wouldn't feel comfortable doing it.
Dworkin: Your worries about the distributional
consequences of organ sales call for regulation of
the market. In Dr. Burrows's system, organs
should be sold only to the state to assure fair dis-
tribution.

Burrows: Transplantation is difficult and differ-
ent from anything else in medicine. I am not say-
ing that because I'm a transplanter. We're not
giving a piece of plastic — a heart valve, or a
pump, or whatever — to the receiver who will re-
ceive it. It's the giving of a vital living organ to
another individual, and it takes a giver and a
receiver, and that's why all these ethical issues
are arising in transplantation. The thing we are
trying to work out is a system to maximize giving
along ethical lines.

Arnold: But selling organs could be wrong for an-
other reason. If I am your friend I might come to
your house and do things for free because I really
like you and you're a really good person. After I
come over you could say, "Here's $10 for your
trouble." That $10 doesn't add something to our
relationship. I may be upset with you because you
are no longer letting me be altruistic.
Baumrin: Okay, so he'll be miffed with the med-
ical community. Is there anything besides his be-
ing miffed that makes it wrong? Some won't be
miffed. Some will be very happy. I was touched
that Dr. Burrows would turn down the Indian
peasant who had one chance at bliss by selling his
kidney. But he was enforcing our view of the
world on the Indian peasant.
Davis: Actually, if you brought the Indian donor
to this country you would have done a lot of good.
The Indians who sell their organs in India are
not having the kind of good consequences that
American organ donors have because they are not
necessarily in as good health, they're not as care-
fully chosen, and they live much more difficult

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

lives. The way it works now, it really hasn't been
the benefit to the poor Indian donors that one
might have expected.

Arnold: Stewart Younger has a wonderful article
in Transplant Proceedings, "The Dark Side of Or-
gan Donation." What he basically says is that ev-
erybody loves organ transplantation. If you ask
the public about it, 99% say they're in favor of it.
But nobody likes thinking about organ procure-
ment. There's a long history of being afraid of
things happening to the dead. It will take some
time to have those superstitions, fears, or what-
ever drives those fears come to the surface and be
worked through. There's a real worry that if it's
pushed too fast, we will basically get the backlash
of those fears and heighten them.

Question: The medical examiner may perform an
autopsy over the objection of the family. If that
law was somehow acceptable in our society, per-
haps presumed consent could also be accepted.
Arnold: That's also true for corneas. In many
states there are presumed consent laws for cor-
neal donations. I don't know how those laws got
passed. But we've been trying to see if there's a
unique legislative history to passing those laws.
There has been some case law involving families
that protested when the corneas were taken with-
out coming to the family to get consent. Courts
have upheld those laws. Yet, for at least the near
future, there will be nowhere near enough im-
plantable organs for the people who would benefit
from them.

General Articles

A Hospital Experience

David Z. Starr, M.D.
Abstract

This article is an account of the author's experience and observations on having been
incorrectly diagnosed as having a myocardial infarction and having been admitted to a
coronary intensive care unit. The article touches on the rigidities of medical thinking, how
one can come to believe what is not true, how to be an "exceptional patient," how medical
care feels from the other side, and other relevant observations.

I RECENTLY had a medical experience that, al-
though certainly not unique, is something many
physicians do not have until later in life, if at
all — an experience whose lessons should be con-
tinually impressed on all physicians. At the age of
43 and slightly overweight, I was admitted to a
coronary intensive care unit because my electro-
cardiogram seemed to indicate I might be having
a myocardial infarction. After five days in two
coronary intensive care units, in which I was
treated as if I had had a myocardial infarction,
three additional days on a cardiac ward, a cardiac
catheterization (which was normal), and a stress
test, it was determined that I had no cardiac dis-
ease, although my cardiogram was somewhat un-
usual for other reasons. The medical conclusion
was that I had had a muscle pull from some vig-
orous karate exercises I had done four days prior
to admission to the hospital.

My theme in this paper, however, is not to
inveigh about the vagaries or misdiagnoses of
modern medical practice. Given the chest pain
and cardiographic abnormality, which could not
be compared to any other previous cardiogram,
the admission was most likely the correct thing to
do. However, I am a psychiatrist and I believe
several observations of what happened to me

The author is in private practice in Brockton and Newton,
Massachusetts; is affiliated with Psychiatric Staff Associates
of Massachusetts General Hospital; and is Instructor in Psy-
chiatry, Boston University School of Medicine. Address re-
print requests to David Z. Starr, M.D., 67 Walnut Hill Road,
Newton, MA 02159.

would be helpful to those engaged in the practice
of medicine.

My observations fall into two categories, one
related to doctors and how they think and the
other to how I felt and was treated as a patient.

How Physicians Think

The most striking observation about how doc-
tors think is the tendency to categorize and then
to continue to operate in accord with the original
categorization. Although my chest pain was atyp-
ical for coronary occlusion (it came on at rest
while I was listening to patients, continued for
seven or eight hours, and was not crushing or
severe or made worse by exertion), once the car-
diogram was read as possible myocardial infarc-
tion, the details regarding the vigorous arm
movements in karate, about which I had told my
doctors, were lost. Even when the laboratory re-
ported that the cardiac fraction of the serum cre-
atine phosphokinase (CPK) level was normal, the
next diagnosis that was entertained was unstable
angina (the laboratory did report that the total
CPK level in the serum was elevated). Nonethe-
less, people kept thinking "heart," and every-
thing was read in this light. I was dealing with
heart specialists, and heart specialists deal with
hearts and think "heart." One doctor, without
even getting all the details, said I probably did
have a myocardial infarction. Only when the car-
diac catheterization was found to be normal did
people begin to step back.

I say this not so much to condemn as to re-
mind us that we all, to some extent, think in cat-

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

Vol 60 No. 1

A HOSPITAL EXPERIENCE— STARR

3gories and tend to generalize from a fixed posi-
tion. We have to be careful to always reevaluate a
diagnosis and not think rigidly. Incidentally, I
live in Boston, and the hospitals involved were
DOth university affiliated, one being a rather
prominent hospital affiliated with a prestigious
Boston medical school.

How Patients Respond

However, the more profound observations
came from my role as patient. On the positive
side, I think the physicians extended themselves
to make sure nothing was wrong because I too
was a physician. However, in their zeal to let me
know I now was a patient and not in control, they
kept me alarmingly in the dark about the techni-
cal aspects of my case. No one had enough time to
show me anything. For example, no one showed
me my cardiogram until I begged it out of the
intern on the night before I was going home, after
everything was determined to be all right — this
despite the fact that the cardiogram was the pri-
mary reason for the misdiagnosis, and all the
tests had been based on this technical item.

Although obviously a sick physician cannot
be his or her own doctor, I think there is an art to
accepting the fact that a physician may need or
want to know more and that there has to be a
balance between this desire for technical informa-
tion and the physician's role as a patient. I think
Aristotle's Golden Mean should be the rule here.
Neither the role of being purely a patient nor let-
ting the physician control his or her own treat-
ment works well. The art lies in balancing the
two, so that some knowledge can be given while
ensuring that the doctor remains a patient.

Another observation I made as a patient is
that it is amazing how easy it is to be persuaded
that you are sick even if you are not. In telling me
that I probably had cardiac disease and treating
me as if I did, replete with all the instrumenta-
tion, nurses, drugs, and so forth, the doctors con-
vinced me for a period that I did have cardiac
disease — so much so that when the first doctor I
saw after the catheterization told me I might not
have any cardiac disease, I did not believe him. I
thought he was wrong. After all, the doctor who
saw me earlier in the day told me I probably had
had a myocardial infarction and he had scared me
into possibly believing him.

It never fails to amaze me how the power of
words and statements from persons in authority
can convince people of things that may not be
true. Although I really was not sick, the caregiv-
ers made me feel I was sick, complete with side

effects from the medications they were giving me
(for example, headache from the nitroglycerine,
which forced me to ask for more medication to
reduce the headache). We all live to a great extent
in our minds, and each of us can be persuaded of
things that are not true, given enough props, re-
gardless of our intelligence or sophistication.

The Exceptional Patient, Still another obser-
vation I made as patient was that to keep my
sanity and not be overwhelmed by what everyone
was doing to me, I had to fight the system to some
extent by not accepting the myriad contradictory
rules of the cardiac intensive care unit and by
questioning each new drug I was to receive with-
out even being asked. I had to be what I later
discovered, when reading Bernard Siegel's excel-
lent book Love, Medicine, and Miracles (1), is the
"exceptional" patient.

By exceptional patients Siegel means those
who fight back, who fight the system, who do not
passively accept their disease or condition, but
who want to know everything that is going on and
to be actively involved in the fight. Siegel, who is
a surgeon who treats many cancer patients and
who has gotten involved with the psychologic as-
pect of illness, divides his ill (usually cancer) pa-
tients into three types:

• those who, either consciously or unconsciously,
want to die to escape the pain of their lives
and/or their illness and who seem to actively
undermine and fight treatment (about
15%-20%)

• the middle of the spectrum — patients who are
compliant and do what the doctor wants (take
their medicine, keep appointments) but tend to
remain passive, expecting the doctor to do the
work; physicians often like these patients be-
cause they are compliant and do not challenge
them (60%-70% of the patients)

• "exceptional" patients — those who actively try
to fight their disease, who want to know the
results of all their laboratory tests, who will
challenge their doctors, who will look into al-
ternate treatments on their own, who are often
considered difficult patients (15%-20%). For
example, Siegel quotes studies to show that
long-term survival is greater in cancer patients
who are judged difficult by their physicians.
That is, they challenge the physicians, they
feel, they get depressed about their illness, they
fight back, they are not stoic.

I found these facts to be true on my own, be-
fore I read Siegel. If you do not fight back to some
extent when fighting back is appropriate, I think
you become even more of a victim. I think this

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

ability to keep some control is what allowed me
not to be traumatized by the whole experience
and what sometimes allows even sicker patients,
as Siegel has observed (1), to keep on living and
improving, even at times with cancer.

How Caregivers Act

I leave two of what I believe are my most
worthwhile observations for last. These have to
do with the twin "k" words: competence and kind-
ness, the two qualities that affected me most dur-
ing my stay. When I was most frightened and in
danger — when I was told that I might be having a
heart attack — I was most reassured by those who
appeared to be competent, who seemed to know
what they were doing, and who seemed also to
care about me. I homed in on this quality the
way a drowning person grabs a log. It is a quality
that some caregivers have and that amazingly
large numbers do not. When you are scared, you
can pick up the difference in a second.

As to kindness, it is a trait we easily forget
about in our pressured, high-technology prac-
tices. But when you are lying in a cardiac inten-
sive care unit at the mercy of your doctors' and
nurses' orientations and whims, believe me — how
kind and flexible they are really counts.

Some nurses showed extra flexibility and

care, even when it meant, say, getting one less
measurement of my blood pressure in the middle
of the night. Among the nurses, I think the lack of
kindness often came out as overrigidity with
rules and, in some cases, covert sadism that one
could sense. These nurses were more involved
with following the rules than really caring for the
patient.

As to the physicians, some, despite their
busyness, also showed an extra bit of kindness
and empathy. But many did not. Among the doc-
tors, the lack of kindness often came out as a cold-
ness, a technical, nonempathic aloofness that,
even if he or she was right, tended to demean me
and make me feel degraded. The patient can eas-
ily pick up what the doctor's attitude to the pa-
tient is, and it makes a big difference.

Perhaps one of the biggest lessons I learned is
that even in our rushed and busy professional
lives, an extra moment of kindness and caring is
felt and appreciated by those on the other end. We
should never forget this. Above all, I learned this:
Always keep an open mind and question your di-
agnoses. No matter how good you are or think you
are, you will sometimes be wrong.

References

1. Siegel B. Love, medicine, and miracles. New York: Harper
& Row, 1988.

Submitted for publication November 1991.

Revision received April 1992.

Equality, Justice, and Liberty
America's Unfinished Agenda

Faye Wattleton

In the mid-19th century, a young man sent his
manuscript of poetry to a man he greatly ad-
mired, Ralph Waldo Emerson. Mr. Emerson read
the work, called Leaves of Grass, and wrote back
to Walt Whitman, "I greet you at the beginning of
a great career."

And so I greet you, the graduates of Mount
Sinai School of Medicine, Class of 1992, at the
beginning of your great careers. By now, you may
think you've had your fill of "good advice," but I
hope you will consider one more bit of advice.

As you pursue your interest in preserving the
health of Americans, I urge you to always keep
your eyes on the bigger picture of life — a concept
illustrated by a story about former Yankee man-
ager Yogi Berra. It seems that one day in Yankee
Stadium, back when streaking was the fad, two
people jumped out of the bleachers stark naked
and rounded all the bases. When Yogi got home,
he told his wife about it, and she asked: "Were
they boys or girls?" Yogi said, "I don't know, they
had bags over their heads."

Yogi missed the bigger picture. And if we fail
to see the bigger picture, we will not comprehend
the vast array of forces aligned against Ameri-
cans' fundamental rights. The bigger picture is
that after 12 years of the New Right agenda and
the systematic reconstruction of the federal
courts, much that Americans have taken for
granted is being dismantled.

The Bush administration's gag rule tells doc-
tors what they can and cannot say to women. In
Planned Parenthood of Southeastern Pennsylva-

Adapted from the author's Commencement Address to the
Class of 1992 of the Mount Sinai School of Medicine on May
11, 1992. At the time of this address the author was immedi-
ate past president of Planned Parenthood Federation of Amer-
ica; she is now developing a nationally syndicated television
show on women's and men's issues.

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

nia V. Casey, the Supreme Court may rule that
doctors become government propagandists; that
women endure waiting periods and husband no-
tification; and that teenagers obtain parental con-
sent before getting abortions. The invasion of the
doctor-patient relationship is a natural extension
of the invasion of women's fundamental rights.
The bigger picture is that both the gag rule and
the Pennsylvania law are attempts to wrest from
women the power to control our reproduction. The
widening circle of oppression that now encroaches
on our first liberty — the right to free speech — also
encroaches on the right of doctors to practice med-
icine.

The notion that the government can censure
our first liberty through the power of the federal
purse strings — and that states can challenge the
courts to overturn our fundamental rights — is re-
pugnant and dangerous for all Americans.
Women face the loss of the fundamental right to
liberty in the controversy over control of our most
private reproductive decisions. And, as has been
the case in the past in this struggle, poor women
suffer first and hardest.

Another aspect of the bigger picture is that
the precious wall between religious dogma and
government policy has been rent. In 1989, in
Webster v. Reproductive Health Services, the Su-
preme Court let stand the preamble to a Missouri
anti-abortion law which says, "The life of each
human being begins at conception." In his dis-
sent, Justice Stevens said the preamble is "an un-
equivocal endorsement of a religious tenet" that
"serves no identifiable secular purpose."

This ever-expanding web of repression en-
compasses the Bush administration's view that it
is appropriate for others to interfere in a woman's
right to freely travel. The Justice Department
joined with Operation Rescue in arguing before
the Supreme Court, in Bray v. Alexandria Wom-

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

(

en's Health Clinic, that a federal statute passed in
1871 to protect blacks from the mob violence of
Ku Klux Klan does not protect women travelling
to abortion clinics from the mob violence of Oper-
ation Rescue. It is an outrage that this adminis-
tration gives aid and comfort to violent mobs that
use the weapon of intimidation against women
and doctors. Our right to travel freely is funda-
mental to our way of life as Americans.

As you have prepared yourselves to serve hu-
manity, you must not lose sight of the threats to
your right to provide those services and the right
of your patients to receive those services.

Each of you has an obligation to actively
work to change the nature of this debate — -to ac-
tively counter the extremists' attempts to cast
women as unconcerned about the consequences of
our sexual decisions.

You can help foster an understanding of the
true nature of America's problems with sexuality
and reproduction — the fact that

• half of the six million pregnancies in this coun-
try every year are unintended

• America has the highest rates of teenage preg-
nancy, childbearing, and abortion in most of
the developed world

• 100,000 women in this country have HIV, and
women are the fastest-growing group of people
with AIDS

• 3,000 children in this country have already
been born with AIDS.

And to counter the extremists' focus on pun-
ishment and oppression, you can be leaders in

challenging the fears about our sexuality, and our
greater comfort with mandating AIDS education
than with mandating sexuality education.

This is the bigger picture that must shape
your resolve and your work. You enter a profes-
sion under attack and in turbulence over these
other issues that are fundamental to the practice
of medicine. Those who are most threatened, the
poor and the young, will depend on your position
of privilege and influence to articulate their
needs and defend their rights.

You have accepted a calling to serve people
without requiring them to be accountable to your
personal moral standard. If you are to meet that
challenge, the political dimension of your profes-
sion must be a part of your work. To ignore that
political dimension — on which you can have a
great impact — is to imperil the health of your pa-
tients, and to render your efforts ineffective
against the forces determined to punish the poor
and to reverse the progress of women.

Political power is never bestowed. It must be
seized — often as the result of a seminal event,
such as the Clarence Thomas hearings. Govern-
ment is not an abstract concept. We, the people,
must control what our government does.

Let our efforts be guided by the words of
Judge Learned Hand: "Liberty lies in the hearts
and minds of men and women. When it dies there,
no Constitution, no law, no court can save it." Let
us heed those words. Let us accept the burden —
and the honor — of keeping liberty alive in our
hearts, so that we may preserve and protect the
health and well-being of all future generations.

Science, Scientists, and Responsibility

Kenneth I. Shine, M.D.

Not long ago, I presented the case of an inter-
esting patient to a group of medical students. He
is a 54-year-old man who had participated in clin-
ical trials of genetically engineered erythropoie-
tin. The drug changed his life. For the previous 14
years, the patient had been on chronic kidney di-
alysis. Although the treatment had adjusted
many of his metabolic parameters, he continued
to be weak, anorectic, and anemic. He could not
work, was sexually impotent, and could not play
with his grandchildren for more than a few min-
utes. Erythropoietin not only largely corrected
his anemia, but treatment was associated with a
dramatic change in his appetite. Indeed, he re-
quired careful counseling in order that the food he
craved not provide more potassium than his rou-
tine dialysis had handled. His strength improved
remarkably so that he could return to work, re-
sume sexual activity, and play with his grand-
children without limit. For him the availability of
this drug was a medical miracle and a triumph of
biomedical science.

Modern Medicine in Microcosm

Several features of this experience deserve
our careful consideration. In the 1950s important
experiments had been done in dog models of ane-
mia which demonstrated that plasma from an
anemic dog could raise the blood count in other
animals. These physiologic experiments, per-
formed in an animal model, ultimately led to the
isolation and purification of the active agent.
During the period in which these experiments
were underway, I had the opportunity as a college

Adapted from the author's address to the annual convocation
of the Mount Sinai School of Medicine (CUNY) on September
16, 1992. From the Institute of Medicine, National Academy of
Sciences. Address reprint requests to the author, President,
Institute of Medicine, 2101 Constitution Avenue, Washington,
D.C. 20418.

undergraduate to take a seminar from a tall, gan-
gling American who had returned from England
a short time earlier, where, with Francis Crick,
he had described the double helix. In a seminar
conducted by James Watson, we speculated about
how long it would take to break the genetic code
and allow the production of important proteins in
vitro. Many of us were convinced that by the end
of the century, the code would be understood and
proteins could be produced. As with so many pre-
dictions, we knew the direction, but had abso-
lutely no sense of velocity.

The development of erythropoietin for clini-
cal use dramatizes the remarkable way in which
various elements of science connect to produce a
clinically useful result. From animal physiology
to molecular biology to pharmacology, various
disciplines of science came together. From the
laboratory to the bedside, the distance has been
shortened both in space and time. As a conse-
quence the products of most fundamental science
reach the patient in extraordinarily short periods
of time. The identification by Slamon and col-
leagues of oncogenes critically related to the nat-
ural history of breast cancer was followed by clin-
ical trials of the protein product within 18
months.

The case dramatizes other aspects of modern
science. The company that now manufactures
erythropoietin was established in collaboration
with university faculty. The early clinical trials
of which this patient was part were carried out by
that company in collaboration with university
faculty, including some at the University of Cal-
ifornia at Los Angeles. By 1988, the investment
by industry in biomedical research had exceeded
that of the public sector. With increasing con-
straints on federal support of biomedical re-
search, in the face of a huge budget deficit, it is
likely that the industrial research component will
grow at a substantially more rapid rate than that
of academic centers. Moreover, many critical

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

January 1993

pieces of equipment, technology, reagents, and
drugs will be in control of industry but will be
essential to the activities of academic scientists.
Since conflict of interest is an important issue of
accountability for contemporary science, and
members of the UCLA faculty participated in
clinical trials, I have avoided any personal, finan-
cial, or other interest in the company that devel-
oped erythropoietin. Erythropoietin is only one of
many important therapies developed out of an in-
dustrial-academic collaboration.

The patient's experience also identifies some
critically important issues in health care deliv-
ery. Last year Medicare spent over $4.4 billion on
long-term kidney dialysis provided to some
58,000 patients throughout the United States.
Treatment of these patients was so profitable that
a variety of free-standing, for-profit enterprises
arose to treat such patients. Erythropoietin and
the drug itself is so expensive that some compo-
nents of the health-care system, for example the
Veterans Agency hospitals, have had a limited
capacity to provide the drug to their patients.
Yesterday's newspapers described the lower uti-
lization of this drug by black patients than by
white patients, a use pattern thought to be par-
tially the result of the large patient copayment
required.

In a recent analysis of health-care costs in
the United States and other industrial nations,
Joseph Newhouse has concluded that the largest
single factor contributing to the high rate of in-
crease of health-care costs in the industrialized
nations has been the diffusion of technology, ac-
counting for approximately 50% of the rate of in-
crease. He and others have pointed out that the
capacity to recapture the investment in technol-
ogy not only in the United States, but also for
companies in Europe and Japan, has been largely
based on the ability to reap substantial product
profits in the American health-care market, and
that control of health-care expenditures in the
United States would have an impact on technol-
ogy development in companies throughout the
world.

In a microcosm, then, this patient's case dem-
onstrates some of the extraordinary features of
modern medical science. Never have the opportu-
nities been greater for discovery and innovation.
The pace of science continues to accelerate, not
only in terms of new discovery, but in communi-
cations between scientists throughout the world
and the capacity of science to take place in other
countries — in industry as well as in the American
university. Results of these discoveries have been
expensive. Developments in industry and the fed-

eral budget could limit the capacity of basic sci-
ence conducted within the university to maintain
America's preeminence in this area.

The American research university is one of
the last remaining areas of this country's inter-
national leadership. But it is threatened. Not
only is direct research support under pressure;
the high cost of tuition and student debt is an
increasing problem. Increasing efforts to control
health-care expenditures limit the extent to
which medical schools can seek support from pa-
tient-care income. The recent emphasis on uni-
versities sharing the cost of research has arisen
out of the budget deficient and the furor over in-
direct cost reimbursements. Confidence has also
been undermined by revelations of scientific mis-
conduct, which is uncommon but has been poorly
handled by the scientific community.

Greater Accountability and
A Larger Social Ethic

All these developments demand a higher de-
gree of accountability by scientists and physi-
cians, accountability for science and for its scien-
tific products. As is well demonstrated by the
erythropoietin story, there is no area of science so
basic that it cannot have a direct, rapid, and in
some cases expensive impact on the society in
which scientists work. As scientists and physi-
cians we must be accountable for our conduct and
that of our peers. The recent David report of the
National Academy of Sciences was notable not
only because of its definition of scientific miscon-
duct, but for the clear statement that scientific
misconduct is real, which is still denied by some,
that it must be recognized, and that it must be
dealt with effectively and expeditiously. In my
judgment the manner in which some elements of
the scientific community responded to scientific
misconduct and the initial response by some to
the issues of indirect cost recovery within re-
search universities undermined the credibility
and the sense of accountability of the scientific
community, which is crucial if scarce resources
are to be provided for science. Too often the argu-
ments for increasing the scientific budget have
been based on the concept that science is valuable
and must be supported in its own right. Yet sci-
entific research is competing with education, so-
cial services, and a variety of other socially desir-
able activities.

Scientists must clearly identify the impor-
tance of what they do in relation to improvements
in health and biotechnology. When new technol-
ogies are developed, both scientists and physi-

Vol. 60 No. 1

SCIENCE AND RESPONSIBILITY— SHINE

cians must ask hard questions about whether the
new development is in fact a significant advance
over what was previously available and will re-
place preexisting technology, or whether it is sim-
ply going to be marginally additive. In this re-
gard, research in health services, with an
emphasis on outcomes and effectiveness of care,
will be a critical component of any reform of the
health-care system and in understanding the
value of new science and technology. Medical
schools must support faculty who can bring
knowledgeable skills to bear on this issue in ev-
ery discipline of medicine.

Accountability also is crucial as we under-
take reform of our health-care system; hard ques-
tions about what is to be done must be asked. We
are increasingly aware of the role of ethics in in-
forming individual decisions made by individual
physicians, patients, and families in the course of
health care. We are also aware of the concept of
professional ethics as it relates to the role of med-
icine and science. But there is a need for a larger
social ethic that addresses the question of what is
right and proper in the way society uses its re-
sources and sets its priorities. The advances in
dealing with chronic kidney disease are remark-
able and often extremely useful. But how do we
balance $4.4 billion for long-term kidney dialysis
with circumstances in which two out of three pre-
school Hispanic children in Los Angeles County
are not immunized, or with the fact that modest
investments in prenatal care would produce not
only healthier babies, but ultimately decrease the
cost of neonatal intensive care? As scientists and
physicians, we must be concerned not only about
the development of new knowledge; we also have
a responsibility for how that knowledge is used
and to whom it is made available.

Finally, scientists have been concerned about
identifying young investigators. There is an in-
creasing awareness of the importance of increas-
ing the size of the pool of scientists from minority
backgrounds as well as enhancing the role of
women in science. From time to time scientists
raise concerns about the pipeline for future scien-
tists. We know that it depends critically on sci-
ence education in the K-12 segment of our edu-
cational system. Led by Bruce Alberts and others,
scientists have made an effort to reach out and
enhance math and science education in the public
schools. A number of medical schools around the
country have had excellent initial experiences,
and one or two have had long-term experience
with math and science education in the public
schools. This is an area in which academic health
centers can show their accountability for both ed-
ucation and science in a dramatic way. There are
a variety of models available in which the facul-
ties and the students in American academic
health centers have created programs in collabo-
ration with teachers of math and science and, in
some cases, with public-school students them-
selves. Academic health centers could provide no
greater community service than to form alliances
in which the considerable energy of faculty and
students could be used to stimulate and to edu-
cate underrepresented students in the public
schools.

Despite all the challenges we face in aca-
demic health centers, we are a privileged commu-
nity., well educated and well compensated. Recog-
nizing and acting on the need for accountability is
ethically and socially essential. In so doing we
increase society's confidence in us and in what we
do, and the long-term result can only be beneficial
for everyone involved.

Jamel L. Oeser, a 16-year-old student at Dr. Martin Luther King High School in New York
City, is enrolled in the Secondary Education Through Health Program (SETH), and is un-
dertaking his research apprenticeship in the Mount Sinai Clinical Microbiology Laboratories,
working directly with Dr. Edward J. Bottone. Jamel is studying the role of the loofah sponge
as a potential public health problem in the transmission of bacteria to the human skin. Jamel's
work has been recognized by the National Institutes of Health, and he is also active in numer-
ous community service projects. He has been writing poetry for about 10 years, mostly while
traveling to and from school on the New York subways.

Heroism

Where black heritage flows from within,
And to be somewhat lighter is as bad as sin.
Take a look at my life.

In my mind I see forests where brown chipmunks
dash,

But through my eyes all I see are rats in the trash.
Take a look at my life.

The bullets are flying. Drugs are easy to sell.
If this is Heaven, I'd hate to see Hell.
Take a look at my life.

Where poverty and hunger are seen as disease,
And all that's ever heard from a policeman is the

cold word freeze.
Take a look at my life.

Where green trees, grass, and mountain tops
Are replaced by abandoned buildings, crack

bottles, and slop.
Take a look at my life.

Where dreams are lost and good things die,
And there's no such thing as apple pie.
Take a look at my life.

Where a young black man has been robbed of his
life.

And the media would rather cover a man's affair

with his wife.
Take a look at my life.

Where the number of unemployed skis down the
slope.

And a young teenage mother can no longer cope.
Take a look at my life.

Where a high-school graduate feels like a fool,
He was a first-rate student in a third-rate school.
Take a look at my life.

An old woman walks by and she clenches her purse.
When asked my opinion, I'm expected to curse.
Take a look at my life.

Where me and my friends are supposed to be thugs.
And I see a good friend losing the battle with drugs.
Take a look at my life.

Where a person educated is labeled a geek.
By "friends" he's tormented because he knows

how to speak.
Take a look at my life.

Where as one teacher put it, slavery's not dead,
The shackles have moved from the wrists to the
head.

Take a look at my life.

Where babies are common but fathers are rare.
They'll all grow up bastards — now how is that fair?
Take a look at my life.

A girl plays in the park, fun and laughter's the sound.
Just then she is shot, her head hits the ground.
Her mother encloses her body, hate sharp as a
knife.

Her tears bathe her girl's body as it starts to lose
life.

Take a look at my life.

Where a man examines his life and lets out a sigh.
He knows there is one way out — that escape is to die.
Take a look at my life.

The Mount Sinai Journal of Medicine Vol. 60 No. 1 January 1993

Critical Care America
is Sensitive to Your
Special Needs

Critical Care America (CCA), an innovative
leader in home infusion therapy, is dedicated to
providing superior quality care and support to
oncology patients in the home setting.

CCA offers a wide range of both traditional
and innovative home infusion therapies, and
complements them with 24 hour expert nursing,
pharmacy, materials management and reim-
bursement services.

We are sensitive to the special needs of the
oncology patient, and have developed a special-
ized program to help meet their needs and the
needs of the physician more effectively and
efficiently than any other home care provider.

Please call us toll-free for more information.
(800) 722-7003 / (800) 344-5500

ffl Critical Care

= AMERICA

Counseling patients about all their
medicines improves their odds of
getting well and staying well.

EVERYONE WINS WHEN YOU TALI

r^end me nfree Medicine Counseling Kit '

Otganizalion

Cily Sljlc /<

Mail to:

National Council on Patient Information

and Education
6f)6 Eleventh Street. NW. Suite HIO
Washington. DC 200(11
To fax your request - (202 1 6.W-(I77^

lis publication
is Qvoilobie in
micfofoftn

Please send me additional information.

University Microfilms International

300 North Zeeb Road 1 8 Bedford Row

Dept. P. R. Dept. P. R.

Ann Arbor, Ml 48106 London, WC1 R 4EJ

U.S.A. England

Name

Institution

Street

City

State

-Zip

■7-.

Because One Size Doesrft Fit All...

New Humulin 50/50 is the tailor-made
answer to individual patient needs. A
unique combination of equal amounts of
Regular human insuhn and NPH human
insulin, it will be useful in situations in
which a greater initial msulin response is
desirable for greater glycemic control.

Like Humulin 70/30t new Humulin 50/50
offers the convenience and accuracy of a
premix. And it can be used in conjunction
with an existing 70/30 regimen.

New soy

Humulin X 50

50% human insulin
isopliane suspension
50% liuman insulin injection
(recombinant DNA origin)

The Newest Option in
Insulin Therapy

WARNING: Any change of insulin should be made cautiously
and only under medical supervision.

" Humulin " 70/30 (70% human insulin isophane suspension,
30% human insulin injection [recombinant DNA origin]).

Global Excellence in Diabetes Care
Eli Lilly and Company

Indianapolis, Indiana
46285

HI-7911-B-249343 c 1992, eli lillv and company

Instructions for Authors

THE MOUNT SINAI JOURNAL OF MEDICINE, established in 1934 as The Journal of the Mount Sinai Hospital, is a peer-re-
viewed general medical journal. It is indexed or abstracted in Index Medicus, Current Contents, Science Citation Index, Hospital
Literature Index, International Nursing Index, Excerpta Medica, Chemical Abstracts, Biological Abstracts, and major databases.
Theoretical and basic science articles for the clinician, reviews, reports of clinical or research experience, articles on any subject
affecting the practice of medicine, short notes in any specialty, book reviews, letters, and articles by medical students are
welcome. The Journal is a participant in the Agreement on Uniform Requirements for Manuscripts Submitted to Biomedical
Journals.

Communications. Address correspondence on manuscripts
review to: Managing Editor, The Mount Sinai Journal of Med-
icine, Box 1094, 19 E. 98th Street, 4D, New York, NY 10029-
6574. Phone: (212) 241-6108; FAX: (212) 722-6386.

Copyright. The publisher reserves copyright and renewal
rights on all published material, and such material may not
be published elsewhere without written permission of the
publisher. When submitting a manuscript, include the fol-
lowing assignment of copyright in your cover letter: In con-
sideration of The Mount Sinai Journal of Medicine (MSJM)
taking action and reviewing and editing my (our) submis-
sion, the author(s) undersigned hereby transfer(s), assign(s),
or otherwise convey(s) all copyright ownership to The Mount
Sinai Journal of Medicine in the event that such work is pub-
lished by MSJM. Each author must sign the statement.

Manuscript Preparation. Submit manuscripts in triplicate,
including three sets of illustrations. Manuscripts must be
typewritten, double-spaced throughout (including refer-
ences, legends, and tables), on one side of 8V2 x 11-in. white
bond paper, with generous margins. Number pages consecu-
tively. Begin a new page for title page, abstract, first page of
text, references, each table, and legends. Computer-generated
typescripts must be double-spaced and must be easy for ed-
itors and typesetters to read. Computer tapes or disks: consult
managing editor.

Title Page. On the title page, on separate lines (double-
spaced), include (1) title of article, phrased as concisely as pos-
sible; (2) name of each author, including first name and de-
gree; (3) name and address of departments and institutions
from which the work originated; (4) acknowledgments of
sources of support; (5) name, place, and date of any confer-
ences at which paper has been delivered; (6) the statement
Address reprint requests to ... , giving full name and ad-
dress, with zip code, of the appropriate author.

Abstract. Original articles should include an abstract, which
should be a single paragraph not exceeding 150 words. State
the purposes; basic approach or methods; main findings; prin-
cipal conclusions. Emphasize new and important aspects.
When an abstract is provided, a summary at the end is not
necessary. Reviews and other articles without an abstract
should conclude with a short summary.

Key Words. Supply 3 to 10 key words or phrases at the bottom
of the title page, for use in indexing the article. Use Med-
ical Subject Headings from Index Medicus.

Writing Style and Abbreviations. Authors should aim for con-
ciseness and clarity without repetition. Avoid medical jargon,
abbreviated phrasing, and obscure or newly coined abbrevia-
tions. Define all abbreviations on first use, except those for
standard, universally recognized units of measurement and
statistical terms. Spell out such terms as white blood cell, he-
matocrit, superior vena cava.

Units of Measurement. Use SI units, giving range or interval
of normal values.

Editing. All papers will be edited to enhance conciseness and
clarity without altering meaning and to insure conformity
with journal style. The author is responsible for all state-
ments in the article.

Illustrations. Photographs, line drawings, graphs, and charts
should increase understanding of the text. Line drawings,
graphs, and charts should be professionally prepared. All il-
lustrations should be submitted in triplicate as sharp, high-
contrast glossy prints. Photomicrographs must be 5Vs in.
wide. On the back of each print affix a gum label with the
number of the figure (numbered consecutively in arabic nu-
merals as cited in text), title of manuscript, name of senior
author, and the word "top" with an arrow indicating top edge.
Illustrations which are to appear together should be mounted
accordingly as plates and should correspond to each other in
size. Protect the prints by enclosing them in heavy cardboard;
do not use paper clips. Photographs of patients can be pub-
lished only if a copy of the permission is submitted with the
photograph. Eliminate names of patients and hospital
numbers from x-rays.

Legends. Legends should not duplicate the text. Double space.
Number each legend to match the illustration it accompanies.
Illustrations mounted as plates to appear together should be
accompanied by a single legend identifying the separate ele-
ments by position ("upper left," and so on) or by letters. If
letters are used in the legend, a corresponding letter must
appear on the print itself Identify all abbreviations. Indicate
magnification and stain for photomicrographs.

Tables. Each table should be typed on a separate page, double
spaced. Number tables consecutively as they appear in text,
using arabic numerals ("Table 1"). Give each table a brief
title (for example, "Effect of DMSO on Rats") and type it on a
separate line. If abbreviations or symbols are used in the
table, identify them in a footnote.

References. When citing references in your text, the first
work cited is numbered 1, and succeeding references are
numbered in sequence; enclose the citation number in paren-
theses and place it before any punctuation: Cytomegalovirus
(1), steroids (2), and recreational drugs (3) have all been im-
plicated (4). The reference list includes only works cited in
the text, numbered in the order in which they are cited. Type,
double spaced, following the general arrangement and punc-
tuation style in the examples below (journal title abbrevia-
tions conform to Index Medicus style):

1. Nadelman RB, Wormser GP. A clinical approach to Lyme
disease. Mt Sinai J Med 1990;57:144-156.

2. International Committee of Medical Journal Editors. Uni-
form requirements for manuscripts submitted to biomedical
journals. N Engl J Med 1991;324:424-428.

When a manuscript is accepted for publication, you will be
asked to confirm the accuracy of all references.

Proofs. Proofs are sent to the corresponding author from the
printer. Stylistic changes which do not alter meaning should
not be made at this stage. Because of increased printing
charges, the cost of excessive author's alterations beyond rou-
tine correction of typographical errors and major errors in
data may be billed to the author. Proofs should be corrected
and returned to the editorial office within 48 hours.

Reprint Orders. A reprint order form is sent to the corre-
sponding author with proofs; return it to the editorial office.

THE

MOUNT SINAI JOURNAL g
OF MEDICINE

Forthcoming:
Grand Rounds

Narrative and Medicine

Oliver Sacks

Albert Einstein College of Medicine

Physician-Writers

Jack Peter Green

Mount Sinai School of Medicine

Osteoporosis: On the Verge of Rational
Effective Therapy?

John Paul Bilezikian
Columbia University College of
Physicians and Surgeons

The Global Impact of Penicillin: Then and Now

Gene H. Stollerman

Boston University School of Medicine

Obscure Gastrointestinal Bleeding

Blair Lewis

Mount Sinai School of Medicine

Pharmacology of Antiinflammatory Agents:
A New Paradigm

Bruce N. Cronstein

New York University School of Medicine

Theme Issues

SEPTE2WER 1993

GUiality Assurance

Editor: Suzanne Kramer

OCTOBER 1993

Cervical Disk Disease:
Controversies in Neurosurgery

Editors: Kal

and Martin H. Savitz

AvanlDle now:

NOVEMBER 1993

Update on Sleep Disorders

Editor: Lee K. Brown

theme section in general issue

Toward the Conquest of Pain

Allan P. Reed, editor
volume 58. no. 3, May 1991
84 pages ♦IS

Festschrift for Rosalyn S. Yalow:
Hormones, Metabolism, and Society

Eugene W. Straus, editor
volume 59, no. 2, March 1992
96 pages SI 5

Modem Management ol
Malignant Melanoma i

Michail Shafir. editor '
volume 59. no 3. May 1992
86 pages 415

Subscriptions for 1993: $65 buHoiduals, US.; $70 all libraries; $75 Individuals outside US.

To order subscriptions or copies of back issues ($15 each), please send check, payable to The Mount Sinai Journal of Medicine, to the Journal at Box 1094, One Gustaoe L Lety
New York, NY 10029-6574: counUr sales at Suite IE. 50 E. 98th Street. New York, NY (phone: 212 241 6108: FAX: 212 722-6386).

\0(im SINAI
F MEDiaNE

JOORNAL

LUME 60 NUMBER 2

MARCH 1993

CONTENTS

GRAND ROUNDS

Osteoporosis: On the Verge of Rational Effective Therapy John

P. Bilezikian 87 ^

The Pathophysiology and Molecular Genetics of Beta

Thalassemia Bernard G. Forget :-95 T?^

Current Concepts of Systemic Necrotizing Vasculitis Lee D. '^_f- ^ c

Kaufman and Allen P. Kaplan ^r/^

CD J-":^'

The Global Impact of Penicillin: Then and Now Gene H. Stol- f>zz

lerman :i32 ^

AUTOPSY AND MEDICAL CARE r-:; j;^^

Editorial: Did the Patient Tell Us? Sherman Kupfer ^ ^

Clinical Correlates of Pathologic Findings: Case 1 Robert Sie-

gel 121

SPECIAL ARTICLES

Narrative and Medicine Oliver Sacks 127

Physicians Practicing Other Occupations, Especially Literature

Jack Peter Green 132

GENERAL ARTICLES

Prevalence of Thyroid Autoantibodies in Ambulatory Elderly Wo-
men Carol Martinez-Weber, Priscilla F. Wallack, Pearl
Lefkowitz, and Terry F. Davies 156

continued inside

Beth

Veterans Israel
Affaire Medical

Center

TfT The Mount Sinai Journal of Medicine is published by The Mount Sinai Medical Center of
I |g New York and has the following affiliates: Beth Israel Medical Center, New York; Bronx

MOUNT SINAI
JOURNAL OF
MEDICINE

Veterans Affairs Medical Center, New York; and Elmhurst
Hospital Center, New York.

Editor-in-Chief

Sherman Kupfer, M.D.

Editor Emeritus

Lester R. Tuchman, M.D.

Associate Editors

Harriet S. Gilbert, M.D. Julius Wolf, M.D.

Managing Editor

Claire Sotnick

Business and Production Assistant

Karen Schwartz

Assistant Editors

Stephen G. Baum, M.D.
David H. Bechhofer, Ph.D.
Constanin A. Bona, M.D., Ph.D.
Edward J. Bottone, Ph.D.
Jurgen Brosius, Ph.D.
Lewis Burrows, M.D.
Joseph S. Eisenman, Ph.D.
Adrienne M. Fleckman, M.D.
Richard A. Frieden, M.D.
Steven Fruchtman, M.D.
Paul L. Gilbert, M.D.
James H. Godbold, Ph.D.

Richard S. Haber, M.D.
Noam Harpaz, M.D.
Dennis P. Healy, Ph.D.
Tomas Heimann, M.D.
Barry W. Jaffm, M.D.
Andrew S. Kaplan, D.D.S.
Samuel Kenan, M.D.
Suzanne Carter Kramer, M.Sc.
Mark G. Lebwohi, M.D.
Kenneth Lieberman, M.D.
Charles Lockwood, M.D.

Lynda R. Mandell, M.D., Ph.D.

Steven Markowitz, M.D.

Bernard Mehl, D.P.S.

Myron Miller, M.D.

Edward Raab, M.D.

Allan Reed, M.D.

Allan E. Rubenstein, M.D.

David B. Sachar, M.D.

Henry Sacks, M.D.

Robert Safirstein, M.D.

Ira Sanders, M.D.

Martin H. Savitz, M.D.
Clyde B. Schechter, M.D.
Michael Serby, M.D.
Phyllis Shaw, Ph.D.
George Silvay, M.D.
Barry D. Stimmel, M.D.
Nelson Stone, M.D.
Max Sung, M.D.
Carl Teplitz, M.D.
Rein Tideiksaar, Ph.D.
Richard P. Wedeen, M.D.

Editorial Board

Barry Freedman, M.B.A.
Richard Gorlin. M.D.
Nathan Kase, M.D.

Panayotis G. Katsoyannis, Ph.D.
Charles K. McSherry, M.D.
Jack G. Rabinowitz, M.D.

John W. Rowe, M.D.
Alan L. Schiller, M.D.
Alan L. Silver, M.D.

Alvin S. Teirstein, M.D.
Rosalyn S. Yalow, Ph.D.

The Mount Sinai Journal of Medicine (ISSN No. 0027-2507; USPS 284-860) is published 6 times a year in January, March, May,
September, October, and November in one indexed volume by the Committee on Medical Education and Publications (Owner), The
Mount Sinai Hospital, Box 1094, One Gustave L. Levy Place, New York, NY 10029-6574. Subscription price: individuals, U.S.,
$65 per year; libraries. $70; individuals outside the U.S., $75. Single copies, $15. New subscriptions begin with the first issue of
the calendar year. Send notice of change of address 60 days before the change is effective. Second-class postage paid at New York,
NY and at additional mailing offices. POSTMASTER: Send address changes to The Mount Sinai Journal of Medicine, Box 1094, One
Gustave L. Levy Place, New York, NY 10029-6574. Copyright 1993 The Mount Sinai Hospital.

-1

Because One Size Doesrft Fit All...

New Humulin 50/50 is the tailor-made
answer to individual patient needs. A
unique combination of equal amounts of
Regular human insuhn and NPH human
insuhn, it will be useful in situations in
which a greater initial insulin response is
desirable for greater glycemic control.

Like Humulin 70/30t new Humulin 50/50
offers the convenience and accuracy of a
premix. And it can be used in conjunction
with an existing 70/30 regimen.

New 50/

Humulin xso

50% human insulin
isophane suspension
50% human insulin injection
(recombinant DMA origin)

The Newest Option in
Insulin Therapy

WARNING: Any change of insulin should be made cautiously
and only under medical supervision.

■ Humulin " 70/30 (70% human insulin isophane suspension,
30% human insulin injection (recombinant DNA origin] ).

Global Excellence iii Dmhetes Care
Eli Lilly and Company

Indianapolis, Indiana
46285

HI-791 1-B-249343 <- 1992. eli lillyand company

Volume 60
Number 2 f
March 1993 |

CONTENTS continued

CASE REPORTS

MRI Documentation of Hemorrhage into Post-traumatic Subdural

Hygroma John 0. Lusins and Ernest R. Levy 161

A Case Report of Neurologic Improvement following Treatment
of Paraneoplastic Cerebellar Degeneration Betty Jane
Mintz and David K. Sirota 163

Index to Advertisers
Eli Lilly and Company page i

Grand
Rounds

Osteoporosis:

On the Verge of Rational Effective Therapy?

John P. Bilezikian, M.D.

I HAVE SUBTITLED this presentation "On the Verge
of Rational, Effective Therapy" because I think
we probably are at that verge for osteoporosis, one
of the most important chronic disorders of life.
When I was just beginning to work in this field,
twenty years ago, one of my mentors said to me,
"There hasn't been a good idea in osteoporosis in
the past 300 years." But over the past twenty
years, the field has changed rapidly; we know
much more about calcium metabolism than we
did, and this knowledge has led to new ideas with
respect to pathophysiology and therapy of osteo-
porosis (1). After a brief introduction to calcium
metabolism, I will concentrate on proposed ther-
apeutic regimens for osteoporosis, an area about
which new information is rapidly becoming avail-
able.

Calcium Metabolism

Three organs, the skeleton, the gastrointes-
tinal tract, and the kidney, interact with a set of
calcium-regulating hormones to keep the human
skeleton sound and the circulating calcium level
within normal limits. An active process involving
the gastrointestinal tract, mediated in part by vi-
tamin D, permits absorption of a certain amount
of calcium. The kidney, under the influence of
parathyroid hormone and phosphate, regulates

Adapted from the author's Grand Rounds in Medicine presen-
tation at the Mount Sinai Medical Center on January 8, 1991.
Final manuscript received September 1991. From the Depart-
ment of Medicine, Division of Endocrinology, Columbia Uni-
versity, College of Physicians and Surgeons, New York. Ad-
dress reprint requests to the author, who is Professor of
Medicine and Pharmacology at Columbia University, College
of Physicians, Department of Medicine 9-410, 630 W. 168th
Street, New York, NY 10032.

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 1993

the amount of 1,25-dihydroxy vitamin D available
to enhance calcium absorption. The skeleton, un-
der the influence of both parathyroid hormone
and 1,25-dihydroxy vitamin D, undergoes con-
stant remodeling. Daily, a substantial amount of
calcium is accreted into bone and resorbed from
bone.

The cells important in the remodeling pro-
cess in bone are osteoblasts and osteoclasts. Os-
teoblasts are responsible for forming the organic
matrix of bone, the osteoid. The mineral phase of
bone, calcium hydroxy apatite, is deposited onto
the osteoid in a manner and under circumstances
that are highly ordered but poorly understood.
Osteoclasts, multinucleated giant cells, are re-
sponsible for resorption of bone, a process that
actually initiates the bone remodeling sequence.
Bone remodeling, the process by which bone is
normally turned over every day, begins with the
osteoclast literally taking a bite out of bone,
thereby excavating a cavity. The results of osteo-
clast activity are associated with an intercellular
signalling event, leading to recruitment of osteo-
blasts, which begin to fill this cavity with new
organic matrix. The organic matrix is then min-
eralized. If everything is in balance, the quantity
of bone removed is matched by the quantity of
bone formed. There is neither gain nor loss of
bone. The dynamic steady state turns over regu-
larly approximately 500 mg of calcium, as much
calcium as is found in two 8-oz containers of milk.

This homeostatic mechanism begins to break
down in all adults such that more calcium is lost
from the skeleton than is gained. The slight im-
balance, which is believed to begin in the mid-
30s, is virtually imperceptible and certainly not
appreciated by routine studies. However, over the
decades, extremely minor daily calcium losses
may become clinically evident in reduced bone

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

Fig. 1. Osteoporosis as seen by scanning electron microscopy. Low-power scanning electron micrographs of iliac
crest biopsies from normal (left) and osteoporotic (right) bone. Note reduction in trabecular plates and their
discontinuity in the osteoporotic specimen as compared to normal biopsy. Adapted from Dempster DW, Shane E,
Horbert W, Lindsay R. A simple method for correlative light and scanning electron microscopy of human iliac crest
bone biopsies. J Bone Min Res 1986; 1:15-21.

mineral density, eventuating in the syndrome os-
teoporosis. As one loses more and more bone, one
is more and more subject to the possibility of frac-
ture at those sites of reduced bone mineral den-
sity. Individuals who have lost bone beyond a cer-
tain amount — the so-called fracture threshold —
are at definite risk for skeletal fracture. The
presence of reduced bone mineral density, how-
ever, does not predict fracture in any one individ-
ual. An intervening event, most likely a fall or
other form of mechanical stress, is required for
the fracture to occur.

In the vertebral spine, the crush fracture of
one vertebra may become associated with other
vertebral crush fractures, leading to the clinical

TABLE 1

Risk Factors Associated with Osteoporosis

Definite risk

Probable risk

Female sex

Sedentary life style

White or Asian origin

Nulliparity

Positive family history

Extreme thinness

Early menopause

High caffeine intake

(or oophorectomy)

High protein intake

Cigarette smoking

Alcohol abuse

Lifelong low dietary

calcium intake

appearance of osteoporosis. Loss of height and a
kyphotic configuration of the back gives a picture
all too commonly seen in postmenopausal women
in this country. Men are not spared this disorder
of the aging skeleton. It begins to become evident
in men approximately a decade later than in
women. We are all likely to develop significant
risk for osteoporotic fracture if we live long
enough.

Osteoporosis is a problem not just in the way
people look, or a problem in back pain, which can
be a source of great morbidity; it is also an enor-
mous economic problem. Osteoporosis is an eco-
nomic problem because of its association with hip
fracture, an event in older women often accompa-
nied by substantial morbidity and mortality. The
osteoporotic hip fracture, a most common cause
for hospitalization in octogenarians, generates a
huge medical bill for this country. Literally, bil-
lions of dollars are spent on health care issues
related to the fractured hip.

Fuller Albright, the father of this field, de-
fined osteoporosis over 50 years ago as "too little
bone," tellingly depicted in Fig. 1. Osteoporosis is
also an architectural problem. Normal bone has
an interconnected honeycombing network that
helps to provide structural integrity. In osteopo-
rosis, there is not only a reduction in bone per

Vol. 60 No. 2

OSTEOPOROSIS— BILEZIKIAN

High

BONE
MASS

Low

FRACTURE
RISK

High

PHASE

CLINICAL
GOAL

Fig. 2. Gain and loss of bone mineral as a function of age. This schematic depiction of skeletal maturation, stability, and
loss is shown in relationship to fracture risk and clinical goals. Adapted from Wasnich RD, Ross PD, Davis JW. Osteo-
porosis: current practice and future perspectives. Trends Endocrinol Metab 1991; 2:59-62.

unit volume but also a loss of connectivity. As
shown in the figure, some trabecular plates are
completely lost, while others are in the process of
losing their base of support (2). It is this problem,
loss of structural integrity, in addition to loss of
bone mineral, that is the challenge in the therapy
of osteoporosis. Somehow, connectivity of trabec-
ular plates must be restored if mechanical
strength is to be regained. Successful therapy,
thus, is not simply a matter of stimulating skel-
etal calcium accretion but a problem in rebuild-
ing normal skeletal architecture.

Risk Factors. Some risk factors for osteopo-
rosis (Table) are well-defined, as for example be-
ing female. White and Asian populations are also
clearly at greater risk than black populations,
which appear to be relatively well-protected. A
positive family history is considered to be an im-
portant risk factor. An early menopause, whether
spontaneous or surgical, clearly places an individ-
ual at greater risk for developing osteoporosis. Al-
cohol abuse is now widely acknowledged to be a
significant risk factor, especially among men.
Cigarette smoking also appears to be detrimental
to the skeleton. Other potential factors such as
lack of exercise, poor calcium diet, caffeine in-
take, and phosphate intake are all under active
investigation.

Achievement of Peak Bone Mass. Does the
problem of osteoporosis begin in childhood and
adolescence? Bone mass is increasing from child-

hood through the early to mid 30s, and then in-
evitably and inexorably declines. The decline in
bone mass is an age-dependent phenomenon, and
occurs without regard to the presence or absence
of risk factors. Achievement of peak bone mass, or
lack thereof, may well be a major point separat-
ing those who become clinically osteoporotic from
those who do not. Achievement of peak bone mass
may well be related to events in one's life when
bone mass is increasing, that is, in the earlier
years before full adulthood is reached. A few ar-
eas of potential relevance in this regard include
dietary intake of calcium as well as other nutri-
tional factors; activity level; weight; menstrual
history; genetic factors (3-5). One must distin-
guish this osteoporosis due to age-related bone
loss (type II osteoporosis) from loss of bone mass
due to estrogen deficiency, so-called postmeno-
pausal osteoporosis (type I osteoporosis). It is im-
portant to recognize that the effects of estrogen
loss are superimposed on age-related bone loss.

Differential Diagnosis of Osteoporosis. The
differential diagnosis of osteoporosis is not simply
a matter of determining whether bone loss is a
phenomenon related to age (made worse by the
presence of risk factors) or to menopause. Osteo-
porosis can be due to a host of endocrine diseases,
such as Cushing's syndrome, hypogonadism, hy-
perthyroidism, or severe primary hyperparathy-
roidism. Corticosteroid-induced osteoporosis is
probably the most common secondary type of os-

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

(

teoporosis. Malignant conditions such as multiple
myeloma and metastatic cancer are well-recog-
nized causes of skeletal demineralization. Immo-
bilization can lead to demineralization. It is a
matter of good medical practice to consider other
causes of osteoporosis, especially in patients for
whom the presentation or severity of the condi-
tion is unusual.

Bone Densitometry. Measurement of bone
mineral density has become an integral part of
the diagnostic evaluation of a patient who is sus-
pected to have osteoporosis. A matter of contro-
versy is the site to monitor. The three major sites
of clinical concern are the vertebral spine, the
hip, and the distal forearm. Although there is a
literature that argues that certain sites in the
skeleton (the ultradistal radius and the calca-
neous) are representative of the skeleton in gen-
eral, these observations are most likely to be
valid only when large population studies are per-
formed. For a given individual, a particular site
might not be representative of other sites and, in
fact, may be misleading. This point is not surpris-
ing considering the fact that the three major sites
are comprised of different proportions of cancel-
lous and cortical bone. Even sites such as the ul-
tradistal radius and the calcaneous, which are en-
riched in cancellous bone, may not be accurate
markers of the cancellous bone in the vertebral
spine. It is now generally agreed that direct mea-
surement of the skeletal site(s) of concern is the
best way to obtain the most accurate information.

A new technology of bone mass measure-
ment, quantitative digital radiography, also re-
ferred to as dual energy x ray absorptiometry, is
rapidly replacing methods such as single and dual
photon absorptiometry (6). This technology is ex-
tremely precise (on the order of 1%) and accurate
(on the order of 3%-5%). Such densitometers can
be used not only for ascertaining bone mineral
density at selected sites but also to monitor
change over time with or without therapy.

Therapy of Osteoporosis

Therapy of osteoporosis has three different
aspects that can be formulated as questions:

• How do we achieve peak bone mass?

• How can we reduce the magnitude of calcium
that is eventually lost from bone?

• Among those who have sustained bone loss,
how can we restore bone mass in such a way
that the skeleton also regains its structural in-
tegrity?

Fig. 2 illustrates these aspects as a function of a
life timeline. To meet these three therapeutic

aims, the list of potential approaches is long.
Some approaches are applicable to more than one
of these aims. I will discuss the following three
general categories: calcium and exercise; hor-
monal agents; pharmacologic agents.

Calcium. "Sticks and stones will break my
bones, and so will too little calcium." Many people
think this cute comment may also be telling in-
sofar as the nutritional value of calcium to the
skeleton is concerned. Put another way, the com-
position of your bones may reflect what you eat.
The well-known Matkovic study compared two
populations that were well-matched except for
their lifelong history of dietary calcium. Subjects
whose diet had been high in calcium had an
incidence rate of hip fracture markedly lower
than those whose diet had been low in calcium
(3). Although studies like this one are few and
far between, they do support the notion that cal-
cium intake in childhood and early adolescence is
instrumental in helping to achieve peak bone
mass.

Whether calcium intake is a critical factor
after the skeleton has completely matured is even
more controversial. A recent study published in
the New England Journal of Medicine (7) showed
that among subjects whose intake was less than
400 mg of calcium, which is indeed a low-calcium
diet, supplemental calcium helped with respect to
spine, femoral neck, and radius bone density, but
individuals whose intake was between 400 and
640 mg did not appear to benefit from supplemen-
tal calcium. This study illustrates that in older
postmenopausal women, a diet obviously defi-
cient in calcium probably is detrimental to main-
taining bone mass.

What is the recommended calcium intake in
this country? The Food and Drug Administration
has raised the recommended daily requirement to
800 mg in the premenopausal state. Most inves-
tigators in the field believe it to be closer to 1,000
mg, the amount of calcium contained in about a
quart of milk. In postmenopausal years, the cal-
cium requirement may increase to 1500 mg.
Where do we get our dietary calcium from? We
get it best from dairy products. Milk, for example,
is an excellent source of calcium. Of course, other
dairy products can also serve as excellent sources
of dietary calcium. With respect to green vegeta-
bles, kale and broccoli contain bioavailable cal-
cium, which is not as true for spinach.

It is difficult to obtain 800 to 1000 mg of cal-
cium through food alone. This is due, in part, to
concerns about calories and cholesterol, which
has limited the intake of dairy products in many
individuals. Parenthetically, low-fat and skim-

Vol. 60 No. 2

OSTEOPOROSIS— BILEZIKIAN

milk products contain as much, if not more, cal-
cium than high-fat dairy products. However, even
persons who have a reasonably normal diet are
unlikely to take in 800 mg. Thus, calcium supple-
ments are usually necessary. There are many
forms of calcium available: calcium carbonate,
calcium citrate, calcium gluconate, calcium lac-
tate, calcium glubionate (8). The particular form
of calcium does not really matter, but calcium
carbonate provides more calcium per weight of
compound than do the other calcium salts because
the carbonate ion is relatively small compared to
the other ions. In terms of cost, some of these cal-
cium preparations, when sold as brand names,
are quite expensive. Generic calcium carbonate
can be very inexpensive. The objective of calcium
supplementation is to reach a total (diet plus sup-
plement) of 800-1000 mg in the premenopausal
years and about 1500 mg in the postmenopausal
years.

Are kidney stones a potential problem with
calcium supplementation? Actually, unless there
is a history of calcium stones, one does not have to
be concerned because it is rare, if not unheard of,
for this to be a complication of supplemental cal-
cium administration.

Exercise. Exercise is an important therapeu-
tic strategy in osteoporosis (9-19). A study from
Australia by Pocock, Eisman, and their associates
indicated that cardiovascular conditioning associ-
ated with exercise, as measured by V02max) was
actually related to bone density of the femoral
neck (11). Thus, conditioning alone could improve
bone mineral density. Questions have been raised
about whether exercise per se is important or
whether in fact the exercise has to be targeted
with respect to general antigravity stress. Anti-
gravity exercises such as walking and running
are, in this respect, considered to be beneficial.
Recent attention has been paid to sites of mechan-
ical stress caused by certain exercises. Tennis
players who stress their dominant arm by serving
have greater bone density of that serving arm
than the other one. Exactly how exercise im-
proves bone mineral density is controversial.
There is no question, however, that exercise is
good and is generally to be recommended. It is to
be recommended at all ages and even, with some
qualification, to those with established osteoporo-
sis.

Exercise is not always beneficial. Certainly
in osteoporotic women who have already sus-
tained compression fractures or in those who have
markedly reduced mineral density, exercise can
be detrimental. If not advised properly on exer-
cise, patients can subject themselves to adverse

musculoskeletal stresses. Another example of the
detrimental effects of exercise is in the premeno-
pausal woman who becomes oligomenorrheic or
amenorrheic due to excessive exercise. These
women are at risk for an estrogen-deficiency type
of bone loss. In a study by Marcus et al., exercis-
ing women with regular periods were compared to
exercising women with amenorrhea (12). There
was a marked reduction in bone mineral density
of the spine in the amenorrheic women. Thus, ex-
ercise must be undertaken in moderation and
with due regard to the importance of maintaining
regular cycles. For most normal purposes, exer-
cise is not going to interfere with menstrual func-
tion unless it is extremely strenuous or associated
with major weight loss.

Estrogen Therapy. The goal of estrogen ther-
apy is to prevent bone loss. Estrogens do prevent
bone loss. The minimal effective dose is 0.625 mg
conjugated estrogen. Estrogens appear to be effec-
tive over the 5-10 year period beyond the meno-
pause. If estrogen therapy is discontinued during
this period, studies have shown that a period of
relatively rapid bone loss may occur, similar to
the loss that is typically experienced in women
who enter the menopause without receiving es-
trogen therapy. Early data suggest that estrogens
may even show some effectiveness well beyond
the menopause (13).

Estrogens are used in a cyclical manner with
a progestational agent. There are contraindica-
tions to the use of estrogens, such as a personal
history of uterine cancer or thromboembolic dis-
ease. A family or personal history of breast cancer
may be another cautionary note, but the data
here are still very controversial. Unfortunately,
many potential estrogen users and their doctors
are generally reluctant to embark on an indefi-
nite period of postmenopausal estrogen therapy
and, thus, it is probably not used as widely as it
should be.

Calcitonin. Calcitonin is an osteoclast inhib-
itor that may be effective in a form of osteoporosis
characterized by high turnover where there is ac-
tive bone resorption and bone formation ongoing
(14). In these individuals, calcitonin seems to be
effective. Whether calcitonin is effective in the
more common presentation of postmenopausal os-
teoporosis, in which bone turnover is very inac-
tive, is not clear.

Fluoride. Fluoride stimulates osteoblasts,
the cell in bone responsible for bone formation. In
many nonblinded studies that have investigated
the use of fluoride in osteoporosis, the results
have seemed to suggest that fluoride is really
quite effective. However, recent well-designed

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

studies have dampened enthusiasm for fluoride in
postmenopausal women with osteoporosis (15).
Using an average of 70-75 mg of fluoride daily,
there was a spectacular increase in vertebral
bone mineral density, up to 35% over a four-year
period. The incremental effect was less dramatic
for femoral sites and not seen at all at the radius.

One question that was critically important to
answer in the study was whether the dramatic
increase in bone mass was associated with a re-
duction in fracture incidence. This is an ex-
tremely important endpoint because an increase
in bone mass does not necessarily mean an in-
crease in structural integrity of bone. The study
showed unequivocally that fracture incidence was
not reduced when fluoride-treated subjects were
compared with placebo-treated subjects.

The other significant outcome of these stud-
ies is the fact that the fluoride group experienced
several adverse side effects, the most noteworthy
ones being dyspepsia, gastrointestinal tract
bleeding, and a lower-extremity pain syndrome.
One can criticize these studies for using a rela-
tively high dose of a short-acting fluoride prepa-
ration and for being relatively short term. But it
is hard not to conclude that fluoride presents real
problems as a therapy for osteoporosis. Any ther-
apy for osteoporosis needs to be associated not
only with improvement in bone mineral density
but also with improved bone strength.

Bisphosphonates. Let us return to our model
of bone remodeling in order to consider a strategy
to overcompensate for the initiation of this pro-
cess by osteoclasts. If osteoclasts are activated
only partially, the excavation cavity will be less
deep. The osteoblast then would fill in this more
shallow cavity with a preprogrammed packet of
bone matrix leading to a net accumulation of
bone. I have described simply the essence of pro-
tocols based on the ADFR concept (Activation,
Depression, Free period, i?epeat). These regimens
have been popularized as "coherence therapy."
Two recent studies have compared bisphospho-
nate therapy with placebo for women with osteo-
porosis (16-17). The regimen consisted of two
weeks of etidronate therapy and approximately
13 weeks of supplemental calcium. Etidronate
was repeated with this same time frame over sev-
eral years. In contrast to the recent fluoride stud-
ies, patients treated with etidronate showed a re-
duction in fracture incidence to accompany a
modest increase in bone mineral density. The re-
duction in fracture incidence was actually greater
among those with the lowest bone mineral den-
sity. Although these reports of bisphosphonate
therapy for osteoporosis are encouraging, it is too

early to know whether they will be upheld over a
longer period of time. One would like to see a
sustained, if not further increase in bone mass,
along with continued reduction in fracture inci-
dence.

Vitamin D. A brief comment on vitamin D
therapy in osteoporosis: There is evidence that
older subjects are deficient in their ability to pro-
duce 1,25-dihydroxy vitamin D due to an age-re-
lated decline in renal function. However, this def-
icit does not appear to be restricted to or worse in
postmenopausal women with osteoporosis. More-
over, most older individuals in this country are
not vitamin-D deficient. Nevertheless, some stud-
ies employing 1,25-dihydroxy vitamin D as a ther-
apy for osteoporosis have shown beneficial effects
(18). Others have not (19). There is a narrow ther-
apeutic window for 1,25-dihydroxy vitamin D; hy-
percalcemia and hypercalciuria can occur depend-
ing on how much 1,25-dihydroxy vitamin D is
used in conjunction with supplemental calcium.
The dose of 1,25-dihydroxy vitamin D and supple-
mental calcium, thus, has to be adjusted carefully
to prevent these adverse effects.

Summary

I would like to leave you with the following
recommendations. In postmenopausal women
without osteoporosis, calcium intake should be
1.5 grams daily, which in almost all such women
requires supplementation of the diet. These
women do not need vitamin D unless it can be
shown that there is a vitamin D deficiency state
present. They should not smoke and they should
not drink alcohol to excess. They should exercise
in a manner appropriate to their general physical
condition. If they can take estrogens, they should
take estrogens.

In women with established osteoporosis,
many of the same suggestions are made with re-
spect to calcium intake, vitamin D, tobacco, alco-
hol, and estrogens. In addition, if there is a high
turnover state in bone — a situation that can be
established only by bone biopsy — calcitonin may
be efficacious. I think it is reasonable at this time
to consider bisphosphonate therapy using regi-
mens that have been published. This is advised,
recognizing that the data are promising at this
time only in the short term. Etidronate should not
be considered for therapy continued beyond two
years until more data are forthcoming.

The last thought I want to leave you with is
that people are going to fall and sustain fractures.
Accidents are going to happen. Even if we were to
realize the most optimistic dream, to prevent os-

Vol. 60 No. 2

OSTEOPOROSIS— BILEZIKIAN

teoporosis in all people, fractures are still inevi-
table. However, when fractures do occur, one
wants to have bones as intrinsically sound as they
can possibly be. The goal, then, is to have a skel-
etal structure that is as healthy as possible, so
that if a fracture were to be sustained, healing
and recovery occurs under the most promising
conditions.

Questions and Answers

Richard Gorlin: Thank you very much, John, for
a spectacular overview. Let me ask you a ques-
tion. Do you distinguish at all between the accu-
mulation of calcium and the necessity to lay down
a proteinaceous matrix first? Is there anything
there that becomes important?
John Bilezikian: That's an important question.
How the osteoid matrix is laid down is probably
important because it is the substrate on which
mineralization occurs. There are some examples
of defects in osteoid formation, but those are usu-
ally congenital defects, associated with another
kind of osteoporosis. In unusual and so far uncom-
mon causes of osteoporosis in adults, there can be
defects in osteoid formation.
Levitt: I enjoyed your talk enormously. I find in-
triguing this disparity between the improvement
in bone density and the perseverance of fractures.
Have data like that been accumulated for estro-
gen? Do you eliminate incidence of fractures with
estrogen as well as increase the bone density?
Bilezikian: With estrogens, bone density is pre-
served but not increased. Fracture incidence is
reduced with estrogens.

Question: Is there a place for bisphosphonate
therapy in patients with multiple myeloma with
extensive bone involvement?
Bilezikian: A good question. There are some data,
not directly related to myeloma but to metastatic
cancer in general. The data are still incomplete. A
drug like a bisphosphonate might be appealing in
the setting of a tumor that has a disposition to
metastasize to bone. We need more information
on this point.

P. Harpel: There's increasing use of estrogen
patches in the treatment of postmenopausal
women. Are there any data on whether that kind
of treatment is effective in preventing osteoporo-
sis?

Bilezikian: The estrogen patch is attractive for
several reasons. The drug administered through
the patch to the skin is estradiol, as compared to
estrone, the principal ingredient in Premarin. Es-
trone must be converted in vivo to estradiol. The
patch eliminates the need for this conversion

step. For this and other reasons having to do with
uptake of estrone in the liver, the patch obviates
potential hepatic problems which occur some-
times with oral estrogens. Another point of note is
that the dose of estradiol administered by the
skin patch is absorbed much more evenly and
slowly than is the oral drug. The problem with
the skin patch is that so far the adhesive used is
associated in a fair number of women with local
skin reactions. With regard to the question of ef-
ficacy, early data do suggest that the skin patch is
effective.

Stein: In your list of predisposing factors, you
listed a high-protein diet. Are you surprised by
that?

Bilezikian: A high-protein diet may be associated
with calcium loss in the urine, but not enough is
known about this point to strongly implicate a
high-protein diet in bone loss.
Gorlin: Thank you very much, Dr. Bilezikian. We
learned a lot.

References

1. Tohme JP, Silverberg SJ, Lindsay R. Osteoporosis. In:

Becker KL, ed. Principles and practice of endocrinology
and metabolism. Philadelphia: JB Lippincott, 1990:
491-504.

2. Dempster DW, Shane E, Herbert W, Lindsay R. A simple

method for correlative light and scanning electron mi-
croscopy of human iliac crest bone biopsies. J Bone Min
Res 1986; 1:15-21.

3. Matkovic V, Kostial K, Simonovic I, et al. Bone status and

fracture rates in two regions of Yugoslavia. Am J Clin
Nutr 1979; 32:540.

4. Ott SM. Attainment of peak bone mass. J Clin Endocrinol

Metab 1990; 71(5):1082A-1082C.

5. Wasnich RD, Ross PD, Davis JW. Osteoporosis: current

practice and future perspectives. Trends Endocrinol
Metab 1991; 2:59-62.

6. Johnston CC, Slemenda CW, Melton LJ. Clinical use of

bone densitometry. N Engl J Med 1991; 324:1105-1109.

7. Dawson-Hughes B, Dallal GE, Krall EA, Sadowski L,

Sahyoun N, Tannenbaum S. A controlled trial of the
effect of calcium supplementation on bone density in
postmenopausal women. N Engl J Med 1990; 323:878-
883.

8. Calcium supplements. The Medical Letter 1989; 31(805):

101-103.

9. Dalsky GP. The role of exercise in the prevention of os-

teoporosis. Compr Ther 1989; 15(9):30-37.

10. Dalsky GP, Stocke KS, Ehsani AA, Slatopolsky E, Lee

WC, Birge SJ. Weight-bearing exercise training and
lumbar bone mineral content in postmenopausal
women. 1988; 108:824-828.

11. Pocock NA, Eisman JA, Yeates MG, Sambrook PN, Ebert

S. Physical fitness is a major determinant of femoral
neck and lumbar spine bone mineral density. J Clin
Invest 1986; 78:618-621.

12. Marcus R, Cann C, Madvig P, et al. Menstrual function

and bone mass in elite women distance runners. Ann
Intern Med 1985; 102:158-163.

13. Lindsay R, Tohme JF. Estrogen treatment of patients

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

with established osteoporosis. Obstet Gynecol 1990;
76(2);291-295.

14. Civitelli R, Gonnelli S, Zaccei F, Bigazzi S, Vattimo A,

Avioli LV, Gennari C. Bone turnover in postmeno-
pausal osteoporosis; effect of calcitonin treatment. J
Clin Invest 1988; 82:1268-1274.

15. Riggs BL, Hodson SF, O'Fallon M, et al. Effect of fluoride

treatment on the fracture rate in postmenopausal
women with osteoporosis. N Engl J Med 1990; 322(12):
802-809.

16. Storm T, Thamsborg G, Steiniche T, Genant HK, So-

rensen OH. Etidronate for postmenopausal osteoporo-
sis. N Engl J Med 1990; 322(18):1265-1271.

17. Watts NM, Harris ST, Genant HK, et al. Intermittent

cyclical etidronate treatment of postmenopausal osteo-
porosis. N Engl J Med 1990; 323(2):73-79.

18. Gallagher JC, Groldgar D. Treatment of postmenopausal

osteoporosis with high doses of synthetic calcitriol. Ann
Intern Med 1990; 113:649-655.

19. Ott SM, Chesnut CH. Calcitriol treatment is not effective

in postmenopausal osteoporosis. Ann Intern Med 1989;
110:267-274.

Grand
Rounds

The Pathophysiology and Molecular
Genetics of Beta Thalassemia

Bernard G. Forget, M.D.

Thalassemia is a hereditary disorder of hemoglo-
bin synthesis (1, 2). Hemoglobin is a tetramer of
two different globin chains: two a and two p
P2). For normal red blood cell metabolism, it is
important to have equal amounts of a and (3
chains, because any excess chains that accumu-
late in the cell are insoluble and precipitate, caus-
ing cellular damage.

Pathophysiology and
Clinical Manifestations

Synthesis of the relative amounts of alpha
and beta chains is measured by incubation of pe-
ripheral blood in the presence of a radioactive
amino acid followed by separation of the globin
chains by column chromatography and quantita-
tion of the radioactivity incorporated into nascent
globin chains. In normal individuals, there is
equal synthesis of a and 3 chains (Fig. 1). The
basic biochemical defect in thalassemia is an im-
balance between a- and 3-chain synthesis, which
can easily be demonstrated by the technique of
isotopic labeling. In patients with homozygous p
thalassemia, only a small amount of radioactivity
is incorporated into newly synthesized (3 chains,
as compared to that incorporated into a chains
(Fig. 1). This disorder, in which some synthesis of
structurally normal (J chains takes place, but in
markedly reduced amounts, is referred to as
thalassemia. In other patients with p thalasse-
mia, there is a total absence of p-chain synthesis,

Adapted from the author's Grand Rounds in Medicine presen-
tation at Mount Sinai Medical Center on June 6, 1990. Final
manuscript received June 1992. From the Hematology Sec-
tion, Department of Medicine, Yale University School of Med-
icine. Address reprint requests to the author, who is Professor
of Medicine and Genetics, Yale University School of Medicine,
333 Cedar Street, PO Box 3333, New Haven, CT 06510.

and that condition is referred to as p^ thalasse-
mia.

Two phenomena result from this imbalance
of globin chain synthesis. First, because of the
quantitative deficit in beta-chain synthesis, there
is reduction in the total amount of hemoglobin
that can accumulate in the cell as a2 P2 tetram-
ers. But secondly, a more important factor in the
pathophysiology of the disorder is the fact that
excess a chains are synthesized and accumulate
in the cell. These excess a chains are the cause of
most of the cellular damage and hematologic ab-
normalities. The peripheral blood smear from a
patient with homozygous P thalassemia demon-
strates two major characteristic findings. First,
the red blood cells are quite hypochromic due to
decreased hemoglobin content as a result of the
deficient P-chain synthesis. But in addition, there
is a great deal of variation in red-cell size and
shape, with red-cell fragments and increased
numbers of reticulocytes and nucleated red cells.
So, in addition to hypochromia due to deficient
hemoglobin synthesis, there is a hemolytic com-
ponent. The accelerated destruction of p thal-
assemic red blood cells results from accumulation
of excess a chains. Although one cannot see them
in regular blood smears, the excess a chains are
visible as inclusion bodies or Heinz bodies when
the blood is examined by phase microscopy (Fig,
2) or by supravital staining.

These inclusion bodies damage the red cells
in a number of ways. The solid nondeformable
physical mass of the inclusion body impairs the
ability of the red cells to navigate through the
microcirculation, in particular the microcircula-
tion of the spleen, where the red cell has to go
through very narrow passages. Unable to go from
the splenic pulp into the venous sinusoids, the
thalassemic red cells become trapped and are de-

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 1993

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

NORMAL

\ \\

1' \l

' w

li A

1 1 1 1

9
I

FRACTION NUMBER

20 30 40

FRACTION NUMBER

Fig. 1. Globin chain synthesis in
human peripheral blood reticulo-
cytes. After incubation in presence of
a radioactive amino acid, globin was
prepared from red cells and frac-
tioned by carboxymethylcellulose col-
umn chromatography. repre-
sents optical density of globin chains;
0--0 represents radioactivity incor-
porated into newly synthesized
globin chains. Top Nonthalassemic
individual; bottom Patient with ho-
mozygous p * thalassemia. Reprinted
with permission from Forget BG,
Kan YW. Thalassemia and the genet-
ics of hemoglobin. In: Nathan DO,
Oski FA, eds. Hematology of infancy
and childhood. Philadelphia: WB
Saunders, 1974, 450-490.

stroyed in the spleen. In addition, the precipitated
a-globin chains damage the red-cell membrane,
either directly or as a consequence of removal or
"pitting" of the inclusion bodies from erythrocytes
by reticuloendothelial cells (Fig. 2).

The process of inclusion-body formation oc-
curs not only in the mature circulating red cells,
but also in the developing nucleated erythroid
precursor cells in the marrow. There is a tremen-
dous amount of destruction of these precursor
cells in the marrow as a result of the membrane
damage caused by inclusion body formation. The
vast majority of cells are destroyed in situ in the
marrow, only a small proportion of cells becoming
circulating erythrocytes. This process is called in-
effective erythropoiesis.

As a result of the ineffective erythropoiesis,
erythropoietin production is stimulated, and tre-
mendous proliferation of the erythroid marrow in
all of the bones of the body takes place. The skull
X-ray of a patient with homozygous 3 thalasse-
mia shows the typical "hair-on-end" appearance
due to thickening of the outer table of the skull,
as well as expansion of the frontal and maxillary
bones. The typical mongoloid facies of patients
with homozygous beta thalassemia is the result of
pronounced erythroid marrow proliferation in the
bones of the skull.

The once typical thalassemic facial deformity
is in fact preventable and is no longer seen in
younger patients because of current improved
treatment. With proper transfusion therapy, one

Vol. 60 No. 2

THALASSEMIA— FORGET

Fig. 2. Phase microscopy of cells from spleen of a patient with homozygous (3 thalassemia.
Inclusion bodies consisting of precipitated a-globin chains (arrows) appear to be removed or "pit-
ted" from a red cell by splenic reticuloendothelial cells (lower left) and are found free in the splenic
pulp (white arrow, lower right). Reprinted with permission from Nathan DG. Thalassemia. N Engl
J Med 1972; 286:586-594.

can prevent the erythroid hyperplasia, because it
is totally dependent on the increased erythropoi-
etin production stimulated by the anemia and in-
effective erythropoiesis. By maintaining a normal
hemoglobin level with high transfusion therapy
one can suppress the stimulation of erythropoie-
sis and prevent these bony changes from occur-
ring. A high transfusion regimen to achieve a
minimum hemoglobin level of 10 g/100 mL is the
currently accepted treatment modality for pa-
tients with transfusion-dependent homozygous (3
thalassemia. However, a consequence of high
transfusion therapy is the development of iron
overload, due to the large number of required
blood transfusions.

Therefore, current therapy consists not only
of high transfusions but, in addition, of active
iron chelation, most effectively by use of the
agent desferrioxamine. Unfortunately this agent
is not effective orally, but must be given paren-
terally, and is most effective when administered
continuously or semicontinuously. Thus, single
intramuscular injections are not as effective as
the current method of administration, which con-
sists of subcutaneous infusion by a motor-driven,
battery-operated pump. Most patients utilize
such pumps twelve hours a day, usually over-
night, for five to six days a week. With this type of

regimen, it is possible to maintain iron balance.
Long-term results of such therapy have begun to
emerge and indicate a beneficial effect in prolong-
ing the life of these patients, who previously
would die in their late teens or early twenties
from cardiac failure or arrhythmias due to iron
overload in the myocardium.

A number of promising iron chelating agents
that are orally effective are currently under
study. It is to be hoped that in the next decade,
some of these oral agents will be introduced into
clinical practice and replace the inconvenient pro-
cedure of subcutaneous desferrioxamine infu-
sions.

Molecular Basis

To understand the molecular basis of p thal-
assemia, knowledge of the structure of the
P-globin gene is essential. The p-globin gene, like
most other genes in our genome, does not encode
its gene product in an uninterrupted linear fash-
ion. The actual protein encoding region of the
gene consists of three coding blocks or exons, sep-
arated by two intervening sequences, or introns.

The process of gene expression is complicated
and includes a number of different steps (Fig. 3).
The gene is initially transcribed as a large pre-

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

/3 GLOBIN GENE

PRE - mRNA

initial transcript

post-transcnptiona I
modifications

nnRNA

GLOBIN CHAIN

-poly A
■poly A

CYTOPLASM

Fig. 3. Schematic representation of process of P-globin gene
expression. Reprinted with permission from ref 2.

cursor molecule containing intronic as well as ex-
onic sequences. This initial transcript is then
modified at both ends: poly (A) is added at the 3'
end and a methylated structure called the cap is
added to the 5' end. The poly (A) confers mRNA
stability, and the cap is important for the inter-
action between the mRNA and initiation factors.
The intervening sequences must then be excised
and the coding sequences precisely ligated to one
another to give the mature messenger RNA. This
process is called splicing, and it is crucial for the
production of functional mRNA. The mature mes-
senger RNA is then transported from the nucleus
to the cytoplasm, where it is translated into a
globin peptide chain. In such a multistep process,
many things can go wrong and, according to Mur-
phy's Law, if anything can go wrong it usually
will. The lesson that we have learned from thal-

assemia is that any step which can conceivably go
wrong has in fact gone wrong in one family or
another with (3 thalassemia. As a consequence,
almost every possible defect in gene expression
that could result in absent or diminished globin
messenger RNA has been described in thalasse-
mia. Thalassemia is the prototype of a genetic
disorder characterized by a relatively homoge-
neous phenotype due to a highly heterogeneous
number of different molecular defects.

Another noteworthy aspect of the process of
gene expression is the existence of a number of
critical sequences in the gene that determine the
processes crucial for proper gene expression. In
the promoter region of the gene, a number of con-
sensus sequences, called boxes, such as the
CCAAT box, CACCC box and TATA box, are im-
portant for the binding of the RNA polymerase
and its associated transcription factors. There
are also important initiation and termination
signals. At the ends of intervening sequences
there are crucial consensus sequences contain-
ing the dinucleotides, GT and AG, that are
the signals for the sites of precise cleavage and
excision of the intervening sequences. Finally, at
the end of messenger RNA, there is another
consensus sequence that is an important sig-
nal for the addition of the poly (A). In such a mul-
tistep process, a small change, such as a single
nucleotide change in one of these critical se-
quences, can have a drastic effect on the accuracy
and efficiency of gene expression and, thus, on the

TIT
TIIT

ITT
Ti

qilr TiTt

ft T

? i

100 bp

R-Globin Gene

J Transcription ^ Frameshift

RNA splicing ^ Nonsense codon

'y' Cap site I Unstable globin

^ RNA cleavage □ Small deletion
y Initiator codon

Fig. 4. Model of human p-globin gene showing sites and types of various mutations causing p thalassemia. Re-
printed with permission from ref 3.

Vol. 60 No. 2

THALASSEMIA— FORGET

Normal Gl^^H^^^^ /3mRNA
Splicing Pathway , , ^globin chain

Fig. 5. Schematic representation of process of alternative splicing in thalassemia due to mutation in IVS-1 at
position 110 (vertical broken line). Reprinted with permission from ref. 2.

quality and quantity of the encoded messenger
RNA.

Over 90 different point mutations of the
(B-globin gene have been reported to cause 3 thal-
assemia (3) by a number of different categories of
defects (Fig. 4), There are mutations in the pro-
moter region of the p gene, in the conserved con-
sensus sequences discussed above, that cause de-
creased transcription of the gene. There are a
number of mutations in the coding blocks or ex-
ons themselves, either nonsense mutations, sin-
gle-base substitutions that change a codon to a
stop signal; or frameshift mutations, one- or two-
base insertions or deletions that change the read-
ing frame of the mRNA and result in premature
chain termination when a new stop signal is en-
countered downstream. One of the more interest-
ing categories of 3 thalassemia consists of the
splicing defects due to mutations, either at the
junctions of the intervening sequences or in the
body of the intervening sequences, that result in
abnormal processing of the precursor mRNA.

Illustrative examples of p thalassemia muta-
tions are provided by molecular defects that are
encountered in Mediterranean individuals. One
such defect is due to a single-base change (C to T)
that converts codon 39 for glutamine (CAG) to a
stop codon (TAG). In this case, the messenger
RNA encodes a truncated protein: instead of a 146
amino acid p-globin chain, translation of the
mRNA results in the synthesis of an abnormal 39
amino acid nonfunctional peptide that is rapidly
turned over and degraded. This (3 39 nonsense
mutation is the second most common cause of
thalassemia in the Mediterranean population.

Another group of mutations are those that
affect splicing. As previously discussed, the in-

variant dinucleotide GT is an important signal at
the beginning of an intron that determines the
precise site of cleavage of the precursor mRNA.
There is a thalassemia mutation, a thalasse-
mia mutation, where a single-base substitution,
changing this GT to an AT, totally blocks the
splicing of the large intervening sequence from
the p pre mRNA; thus no normal p globin mRNA
is produced and there is total absence of p globin
chain synthesis.

The most common form of p thalassemia in
the Mediterranean basin has been shown to be
due to an interesting and novel mechanism for
abnormal gene expression: the creation of a new
splicing signal in an intron that results in the
processing of the majority of the p pre mRNA
transcripts into abnormal mRNA molecules, only
a minority of the pre mRNAs yielding normal P
mRNA. The mutation is located at nucleotide 110
of IVS 1,21 nucleotides upstream from the nor-
mal 3' or acceptor AG of the first intron of the
P-globin gene and consists of an A to G base sub-
stitution. This mutation creates a new acceptor
AG, some 20 nucleotides upstream of the AG that
is normally used, and in a sequence context that
is similar, if not identical, to the normal 3' splice
site. As a result of this base change in an appar-
ently unimportant part of the gene, a new signal
is created that appears to be preferred over the
normal signal by the splicing mechanism. The
mutant pre mRNA can be spliced in one of two
ways. It can be spliced normally; but this happens
in only the minority of cases, approximately 10%
of the time. For some reason, the new cryptic
splicing signal created by the mutation is pre-
ferred and is used 90% of the time, giving a mu-
tant messenger RNA that contains a small insert

100

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

of intronic sequence (19 nucleotides long) that has
a stop signal in it so the mRNA can only encode a
truncated, abnormal p globin chain (Fig. 5).

In summary, studies of the molecular basis of
thalassemia have demonstrated that the molecu-
lar cause of thalassemia is quite heterogeneous.
Over 90 different mutations have been identified,
yet the resulting phenotype is clinically and bio-
chemically rather similar in most cases. The end
result is the absent or decreased synthesis of p
globin chains. As a consequence of that deficiency
of 3 globin chains, excess a chains accumulate,
damage the red cell, and lead to the clinical pre-
sentation of hemolytic anemia and ineffective
erythropoiesis with marked marrow expansion.
Despite the vast heterogeneity of molecular de-
fects, the biochemical and clinical phenotype is
quite similar because of the common denominator
of absent or decreased synthesis of (3 globin
chains, with resulting accumulation of excess a
globin chains that cause the damage to the af-
fected erythroid cells.

A number of additional general conclusions
can be drawn from the results of studies on the
molecular basis of p thalassemia, and these con-
clusions have implications for the prenatal diag-
nosis and future prospects for gene therapy of p
thalassemia. A given mutation is generally found
only within one racial group and not another.
Thalassemia is quite common, not only in Medi-
terranean populations, but also in Afro- American
and Asian populations. However, in these differ-
ent racial groups one usually finds a different set
of specific mutations. A given mutation also tends
to be on the same chromosomal background in
different individuals of the same racial group. If
one determines polymorphisms in and around a
P-thalassemic globin gene, one then has a clue to
what mutation is likely to be present in that gene.
Finally, a small number of mutations usually ac-
count for the majority of cases of p thalassemia in
a given population. For instance, despite the over-
all high number of different p thalassemic muta-
tions, over 90% of all the mutations observed in
Mediterraneans can be accounted for by only six
different molecular defects, the two most common
being the nonsense mutation at codon 39 and the
IVS 1 position 110 alternative splicing defect.

Prenatal Diagnosis

Information on the molecular basis of p thal-
assemia has been translated into new, highly
effective, efficient, and accurate techniques for
prenatal diagnosis. Initial procedures and ap-

proaches for the prenatal diagnosis of p thalas-
semia had to rely on obtaining fetal blood samples
because the abnormal biochemical manifestation
is restricted to red blood cells. This was usually
carried out in the mid trimester by the procedure
of fetoscopy, in which a needle was inserted into
the uterine cavity and used to aspirate small
amounts of blood from fetal blood vessels for anal-
ysis of globin chain synthesis. Although accurate
and effective, this approach is not totally safe.
Even in the best of hands, fetoscopy has a mor-
tality and morbidity rate of approximately 5%.
Knowledge of the nature of the p thalassemic mu-
tations at the DNA level has now made possible
the use of fetal DNA rather than fetal blood for
the purpose of prenatal diagnosis. The source of
fetal calls may be amniotic fluid obtained by reg-
ular amniocentesis in the mid trimester, or, as is
now preferred, chorionic villi obtained by trans-
cervical biopsy in the first trimester.

A technique that has revolutionized the abil-
ity to study mutations for any genetic disorder in
small amounts of DNA is the polymerase chain
reaction (4). The basic principle is to make syn-
thetic oligonucleotide primers approximately
twenty bases in length to either side of a region of
interest of the gene, and to submit the DNA to
repeated cycles of denaturation and in vitro rep-
lication by DNA polymerase using the primers as
the site of initiation of DNA synthesis. One thus
obtains geometrical amplification of the region of
interest with the production, after 30 to 40 cycles,
of over a million copies of the target sequence.
Thus, starting with only 1 |xg of total cellular
DNA, one can obtain 1 |jLg of globin (or other) gene
fragment for analysis.

There are two basic approaches for analyzing
such amplified DNA to detect the presence of sin-
gle-base mutations. One method is illustrated by
the disorder in sickle cell anemia, in which there
is a single-base substitution in codon number 6,
changing a glutamine codon (GAG) to a valine
codon (GTG). This single-base change also hap-
pens to change the recognition site for a restric-
tion endonuclease, that is, an enzyme that cleaves
DNA at a specific sequence. In normal DNA,
there is a cleavage site for the enzyme Mstll that
encompasses codon 6. In the sickle cell gene, that
particular recognition sequence is abolished, so
when mutant DNA is digested with the enzyme, a
larger fragment is obtained than that obtained
with normal DNA. If one amplifies this region of
DNA as a small fragment of approximately 300
base pairs (bp), one can then test for the ability of
that piece of DNA to be digested by Ms^II. The

THALASSEMIA— FORGET 101

Vol. 60 No. 2

DNA fragment from a normal individual is com-
pletely cleaved into two subfragments of approx-
imately 200 and 100 bp respectively. The DNA
fragment of a patient who is homozygous for
sickle cell disease remains totally undigested.
Amplified DNA from an individual who is a het-
erozygous carrier for the sickle cell gene gives a
pattern in which half of the DNA is uncleaved,
and half is cleaved into the two smaller subfrag-
ments. Approximately 30% to 40% of the ^ thal-
assemia point mutations can be directly identi-
fied by the alteration of a restriction endonu-
clease site in amplified DNA.

However, the majority of mutations do not
affect restriction sites; an alternative technique is
available to detect such disorders. The approach
consists of hybridization of the amplified DNA to
allele-specific oligonucleotides (ASOs). The proce-
dure uses two synthetic oligonucleotide probes
that overlap the site of the mutation, one identi-
cal to the mutated sequence, the other to the nor-
mal sequence. The single-base difference in these
synthetic oligonucleotides, —20 bases in length, is
located in the middle of the sequence and condi-
tions can be established so that each oligonucle-
otide will hybridize only to its identical comple-
mentary sequence, and not to the sequence that is
one base different. Amplified DNA from a normal
individual will hybridize only to the normal probe
and not to the mutant probe. DNA from a ho-
mozygous p thalassemic individual will hybridize
only to the thalassemic probe and not to the nor-
mal probe. On the other hand, amplified DNA
from an individual heterozygous for the p thalas-
semia mutation will hybridize equally well to
both probes. Therefore, the technique of allele-
specific oligonucleotide hybridization allows the
detection of either homozygosity or heterozygos-
ity for a given mutation and can be readily ap-
plied to the analysis of amplified DNA fixed to a
filter.

The advantage of chorionic villus biopsy as a
source of fetal cellular DNA is that the procedure
can be done early in pregnancy, at eight to ten
weeks of gestation, which is a much more accept-
able time for family planning. The couple can
make a decision much earlier in the pregnancy
rather than having to wait until the second tri-
mester, which is the time when amniocentesis is
usually done.

In summary, the implications of molecular
studies for prenatal diagnosis of p thalassemia
are the following: DNA-based diagnosis, which is
safer than fetoscopy, is feasible in the vast major-
ity of cases, and has been greatly facilitated by

the advent of the polymerase chain reaction. The
procedure does require knowledge of the specific
molecular defect in the parents, but with the
speed and efficiency of the polymerase chain re-
action, it should be possible to rapidly determine
the mutations in a given couple at risk and
thereby establish the appropriate procedures for
prenatal diagnosis. DNA-based prenatal diagno-
sis of p thalassemia and other hemoglobinopa-
thies is a current reality and is a direct extension
and application of studies on the molecular basis
of these disorders.

Prospects for Gene Therapy

The implications of studies on the molecular
basis of p thalassemia for gene therapy of the dis-
order can be summarized as follows. A single ef-
fective approach to gene therapy would be the in-
troduction of a functional p-globin gene into
hematopoietic stem cells of the homozygous af-
fected individual as a substitute for the defective
genes. The transfer of a normal p-globin gene into
the hematopoietic stem cells of an affected pa-
tient's bone marrow, and reinfusion of that trans-
fected or transduced marrow, is one potential ap-
proach to gene therapy (5, 6).

A number of obstacles must be overcome be-
fore gene therapy becomes a reality. First, one
needs to have regulated expression of the trans-
ferred gene. When one transfers a globin gene
into hematopoietic stem cells, one wants it to be
expressed only in the erythroid cells and not in
the nonerythroid progeny of the stem cells. Also,
one needs to have high levels of expression of the
transferred gene. The amount of hemoglobin
made in the red cell is considerable, and to ade-
quately replace the deficient p-globin chain syn-
thesis, one has to have very high levels of expres-
sion of the transferred globin gene. Finally, the
gene must be transferred into the self-renewing
or reconstituting stem cell in order to obtain long-
lasting therapeutic results. A number of experi-
ments indicate that if one transfers a human
P-globin gene into the hematopoietic stem cells of
a mouse, it will be expressed essentially only in
erythroid cells and not in nonerythroid cells (7).
Even though the transferred gene becomes inte-
grated in a chromosomal environment different
from that where it normally resides, it contains
sufficient sequence in its immediate flanking
DNA to confer erythroid cell-specific expression.
Therefore regulated expression of transferred
globin genes is not a real problem. On the other
hand, the levels of expression of transferred

102

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

globin genes have been disappointing. Experi-
ments in mice (7) and in tissue culture cells usu-
ally result in only approximately 5% to 10% as
much expression of the transferred gene as that of
the endogenous globin gene. Such a level would
not be enough to reverse the defect in p thalasse-
mia. One needs to achieve a level of expression
from the transferred globin gene(s) that is nearly
as high as that of a normal endogenous globin
gene.

Recently, there has been important progress
in the development of strategies to increase the
amount of expression of transferred globin genes
to levels that would be therapeutically effective.
The advance that should contribute to achieving
high levels of globin gene expression following
gene transfer is the recent discovery of interest-
ing sequences in the p-globin gene cluster that
are located far upstream from the globin genes
and were initially identified as hypersensitive
sites to digestion by DNase I (8). The importance
of these sequences was first suggested by the find-
ing of a number of mutations in which the
P-globin gene itself is structurally normal but not
expressed when located on a chromosome that
carries a deletion of these sequences. Four such
deletions have so far been identified and are as-
sociated with the phenotype of 78(3 thalassemia.
These deletions remove different amounts of the
P-globin gene cluster, the most striking being
Hispanic 783 thalassemia, in which all the
P-like globin genes are intact but silent due to a
—30 kb deletion involving three of the four hy-
persensitive sites (9). When DNA sequences con-
taining these DNase I hypersensitive sites are
linked to a p-globin gene that is used to produce
transgenic mice, the level of expression of the
transferred (B-globin gene is virtually equal to
that of the endogenous gene (8). These sequences
have been called the LCR, or locus control region.
The LCR confers high levels of erythroid-specific
expression to transferred 3-globin genes, in a
copy-dependent and position-independent man-
ner, that is, the level of expression is proportional
to the number of copies of the gene that are trans-
ferred and is uniformly high irrespective of the
chromosomal site of integration of the transferred
gene(s). The DNA sequences that are critical for
the functional activity of the LCR have been lo-
calized to a small region of DNA surrounding the
DNase I hypersensitive sites, so it should be pos-
sible in the future to insert these sequences into
retroviral (or other) gene therapy vectors to ob-
tain near normal levels of expression of trans-
ferred p-globin genes.

The most likely approach to gene therapy for
P thalassemia will involve the use of retroviral
vectors because of the efficiency with which such
vectors can introduce genes into hematopoietic
(and other) cells (5, 6). One concern with such
vectors is safety and the possibility that the dis-
abled vector DNA initially rendered incapable of
producing infectious viral particles could recom-
bine with normal viral sequences and lead to
chronic productive viremia, with the possibility of
insertional mutagenesis leading to malignant
transformation. However, a number of systems
have been developed that greatly reduce the
chance of such viral recombination, although in-
sertional mutagenesis is still remotely possible
with only a single cycle of retroviral introduction
and chromosomal insertion into hematopoietic
stem cells. Another issue in the use of retroviral
vectors is the feasibility of retroviral transfer into
the pluripotential or reconstituting hematopoiet-
ic stem cells. Because retroviral integration re-
quires one cycle of cell division, and pluripoten-
tial stem cells are predominantly in the resting or
Gq phase of the cell cycle, the frequency of retro-
viral infection (or transduction) of pluripotential
stem cells is much lower than that of more ma-
ture, and therefore shorter lived, hematopoietic
progenitor or precursor cells. However, the pre-
stimulation of target marrow cells with various
combinations of hematopoietic growth factors has
greatly increased the frequency of successful stem
cell transduction (10).

The general approach to gene therapy would
consist of the aspiration of bone marrow from the
affected patient followed by the in vitro exposure
of the marrow cells to defective retroviral parti-
cles, containing a copy of a normal p-globin gene,
that have been modified so their genetic material
(RNA) lacks the necessary information to permit
viral replication. After entry into the marrow
cells, the reverse transcriptase enzyme of the vi-
ral particle will make a DNA copy of the retrovi-
ral genome, which will then become integrated in
a random fashion within the chromosomal DNA
of the marrow cells. The transduced cells are then
reinfused intravenously into the patient, where
they will hone to the bone marrow space and con-
tinue to proliferate normally. The transferred
P-globin gene should be expressed exclusively,
and it is hoped at a high level, in the erythroid
cell progeny of the transduced stem cells.

Although actual attempts of long-term (that
is, stem cell) gene therapy for (3 thalassemia are
many years away, short-term gene therapy for
another genetic disorder, adenosine deaminase

Vol. 60 No. 2

THALASSEMIA— FORGET

103

deficiency, has already been initiated by means of
repeated transduction and reinfusion of periph-
eral blood lymphoid cells (6).

Summary and Conclusions

Review of the pathophysiology and molecular
basis of P thalassemia reveals that an extremely
heterogeneous group of molecular defects can
give rise to a relatively uniform clinical and he-
matological phenotype that is primarily the re-
sult of the excess of free a-globin chains that ac-
cumulate in the face of absent or markedly
reduced p-globin chain synthesis. Despite the mo-
lecular heterogeneity, it has been possible to es-
tablish highly accurate and efficient DNA-based
prenatal diagnosis for p thalassemia. Important
progress is also being made in the area of gene
therapy for p thalassemia.

References

1. Weatherall DJ, Clegg JB. The thalassemia syndromes,

3rd ed. Oxford: Blackwell Scientific Publications, 1981.

2. Bunn HF, Forget BG. Hemoglobin: molecular, genetic and

clinical aspects. Philadelphia: WB Saunders, 1986.

3. Kazazian HH Jr. The thalassemia syndromes: molecular

basis and prenatal diagnosis in 1990. Semin Hematol
1990; 27:209-228.

4. Eisenstein B. The polymerase chain reactions: a new

method of using molecular genetics for medical diagno-
sis. N Engl J Med 1990; 322:178-183.

5. Karlsson S. Treatment of genetic defects in hematopoietic

cell function by gene transfer. Blood 1991; 78:2481-
2492.

6. Anderson WF. Human gene therapy. Science 1992; 256:

808-813.

7. Dzierzak EA, Papayannopoulou T, Mulligan RC. Lineage-

specific expression of a human p-globin gene in murine
bone marrow transplant recipients reconstituted with
retrovirus-transduced stem cells. Nature 1988; 331:35-
41.

8. Townes TM, Behringer RR. Human globin locus activa-

tion region (LAR): role in temporal control. Trends in
Genetics 1990; 6:219-223.

9. Driscoll MC, Dobkin OS, Alter BP. -/Sp-Thalassemia due

to a de novo mutation deleting the 5' p-globin gene
activation-region hypersensitive sites. Proc Natl Acad
Sci USA 1989; 86:7470-7474.
10. Bodine DM, Karlsson S, Nienhuis AW. Combination of
interleukins 3 and 6 preserves stem cell function in
culture and enhances retrovirus-mediated gene trans-
fer into hematopoietic stem cells. Proc Natl Acad Sci
USA 1989; 86:8897-8901.

Grand
Rounds

Current Concepts of Systemic
Necrotizing Vasculitis

Lee D. Kaufman, M.D., F.A.C.P. and Allen P. Kaplan, M.D.

The primary idiopathic vasculitides are gener-
ally distinguished from one another by the size of
the vessel involved, the organ systems affected,
and the pathologic presence or absence of a gran-
ulomatous lesion (1). Descriptive classifications
(outlined briefly in ruled box) based on these fea-
tures have been the subject of previous reviews
(2-4). Since the classic original description of
periarteritis nodosa by Kussmaul and Maier (5),
it is now appreciated that notable overlap exists
within the clinicopathologic spectrum of systemic
vasculitis (6). The preferable term, systemic nec-
rotizing vasculitis (SNV), refers to a group of dis-
orders characterized by an inflammatory infil-
trate in the wall of small to intermediate-sized
muscular arteries associated with clinical disease
from resultant ischemia or infarction of the kid-
ney, gut, nerve, myocardium, lung, skin, and gen-
itourinary tract (6).

The histopathology of SNV evolves over time.
Early lesions are characterized by a panarteritis
with infiltrating polymorphonuclear leukocytes
(PMN), fibrinoid necrosis, endothelial cell injury,
and thrombosis. Mononuclear cells with intimal
proliferation t5TDify the chronic stages. There is a
predilection for vascular injury to occur with a
focal distribution, often at sites of arterial bifur-
cation, producing microaneurysmal dilatation
(7, 8).

Adapted from a presentation by APK at Mount Sinai Medical
Center on May 5, 1990. Final manuscript received June 1991.

From the Division of Allergy, Rheumatology, and Clinical
Immunology (LDK) and the Department of Medicine (APK),
The State University of New York at Stony Brook, Stony
Brook, NY. Address reprint requests to Lee D. Kaufman,
M.D., F.A.C.P., Division of Allergy, Rheumatology, and Clin-
ical Immunology, Health Sciences Center, The State Univer-
sity of New York at Stony Brook, Stony Brook, NY 11794-
8161.

Etiopathogenesis

The current understanding of the pathogen-
esis of SNV is based on the serum sickness model
of immune complex disease in both animal mod-
els and humans (9, 10). In this paradigm, circu-
lating immune complexes form in the presence ol
antigen excess and, modulated by physical factors
such as local temperature, turbulence, size, anc
electrical charge, deposit in the vascular wall (9
11). In most forms of vasculitis the "triggering'
antigen is unknown; however, the association ol
specific infectious agents, neoplastic disorders
and drugs with SNV provides insight for the
pathogenesis of these predominantly idiopathic
syndromes. Hepatitis B has been noted to occur ir
from 20% to 30% of patients with polyarteritis
nodosa (12). The use of more sensitive assays thai
employ monoclonal antibodies against the hepa-
titis B surface antigen, or molecular probes foi
viral DNA in the peripheral blood lymphocytes oi
affected patients, has recently been reported tc
increase the identification of hepatitis B infectior
in vasculitis patients who were previously sero-
negative with standard polyclonal assays (13)
Using these techniques, the prevalence for hepa-
titis B seropositivity in polyarteritis was 75*%
(13). Additional evidence for infectious agents has
included SNV in individuals with c3d;omegalovi-
rus (14), parvovirus (15), human immunodeficien-
cy virus-1 (16), varicella-zoster (17), Epstein-Ban
virus (18), rubella (19), Borrelia burgdorferi (20).
the recently identified hepatitis C (21), and SNV
as a sequela of acute serous otitis media (22).

The best model of neoplastic disease associ-
ated with SNV is hairy cell leukemia (23). Vas-
culitis has also occurred in association with car-
cinoma of the colon, nasopharynx, prostate,
kidney, breast, cervix, and lung; melanoma; and
pheochromocytoma (24). Penicillins, sulfon-

104

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 1993

Vol. 60 No. 2

NECROTIZING VASCULITIS-KAUFMAN & KAPLAN

105

Classification Schema for Vasculitis
I. Large vessel arteritis
Takayasu's arteritis
Giant cell arteritis
II. Medium and small vessel systemic necrotizing
vasculitis (SNV)

Polyarteritis nodosa (PAN) group
Classic PAN

Churg-Strauss (allergic granulomatosis)
"Overlap"

Microscopic polyarteritis

Kawasaki disease (infantile PAN)

Amphetamine induced SNV

SNV associated with Crohn's disease

SNV associated with infection

SNV associated with neoplasm

SNV associated with the connective tissue

disease
Localized PAN

Skin

Appendix
Gallbladder
Kidney
Lung

Mesentery

Uterus

Breast

Epididymis

Testis

Wegener's granulomatosis
Lymphomatoid granulomatosis
III. Small vessel arteritis/venulitis

Hypersensitivity vasculitisAeukocytoclastic
vasculitis
Infection

Drug/serum sickness
Neoplasm

Connective tissue disease
Essential mixed cryoglobulinemia
Henoch-Schonlein purpura

amides, phenytoin, and heterologous antiserum
(most recently antithymocyte globulin) are
among the most important causes of drug-related
vasculitic disease (4, 10).

Complement activation by in situ immune
complexes generates cleavage fragments with im-
portant biologic activities that mediate chemo-
taxis of PMN (C5a) and degranulation of baso-
phils and mast cells (anaphylotoxins C3a, C4a,
and C5a). Histamine release by these fragments,
histamine-releasing factors from infiltrating
mononuclear cells (25), or potentially anti-IgE
autoantibodies (26) promote localization of im-
mune complex material along the endothelial
basement membrane (27). Further recruitment of
PMN is likely to be augmented by leukotriene-B-
4, platelet-activating factor, and the neutrophil-
activating peptides-1 and -2 (28, 29). Cytokines
elaborated by infiltrating mononuclear cells in-
crease the expression of adhesion proteins on the
endothelial surface (30). These include endotheli-
al-leukocyte adhesion molecule- 1, which is in-
duced after stimulation with interleukin (Il)-l
and tumor necrosis factor (TNF)-a, and intercel-
lular adhesion molecule- 1, present on resting en-
dothelial cells and up-regulated by II- 1, TNF-a,
and interferon-7.

Both humoral and cellular immunity appear
to be important in the generation of pathologic
variants of SNV. Granuloma formation may be
related to a T-cell response to autoantigens and
cytokine production. In that regard, Il-l, TNF-a,
11-6, and interferon-7, which are influential in the
generation of granulomas, presumably play a piv-
otal pathogenetic role in Churg-Strauss vasculitis
and Wegener's granulomatosis (31). Further-
more, 11-5, crucial to eosinophil growth and acti-
vation, may be critical to the eosinophilia and tis-
sue injury of Churg-Strauss vasculitis (32).
Granulocyte-macrophage colony-stimulating fac-
tor, a product of T cells and macrophages, has
recently been observed to produce lesions of nec-
rotizing vasculitis at sites of cutaneous injection
(33). 11-4 activates monocytes and macrophages to
become giant cells and may be central to the pro-
duction of vasculitis-specific antineutrophil au-
toantibodies (see below).

As PMN are recruited by chemoattractants
and secured to the endothelium by adhesion mol-
ecules, they release proteases and oxygen radicals
that degrade basement membrane components
and produce vascular injury. The exposure of
basement membrane permits activation of the
clotting and fibrinolytic pathways via Hageman
factor (34). The presence of Il-l and TNF-a, which

increase tissue thromboplastin and decrease plas-
minogen activator release from endothelial cells,
creates a prothrombotic milieu favoring occlusive
disease (35, 36).

Clinical Disease

The majority of patients with SNV will have
constitutional features such as fatigue, weight
loss, fever, arthralgias, and myalgias. The preva-
lence of specific target organ involvement varies
with each vasculitic syndrome. Necrotizing vas-
culitis can be subdivided into a number of syn-
dromes, all characterized by clinical overlap: clas-
sic polyarteritis nodosa (5); the Churg-Strauss
variant of allergic granulomatosis in individuals
with a background of atopic disease, asthma, and
eosinophilia (37); overlap vasculitic syndromes

106

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

(6); microscopic polyarteritis with prominent re-
nal disease that resembles that seen in Wegener's
granulomatosis (38); vasculitis associated with
amphetamine abuse (4); necrotizing vasculitis
with Crohn's disease (39); Wegener's granuloma-
tosis; and the systemic vasculitis associated with
connective tissue or immunologically mediated
disorders. The latter include systemic lupus
erjrthematosus, rheumatoid arthritis, Sjogren's
syndrome, Behcet's disease, relapsing polychon-
dritis, Cogan's syndrome, Buerger's disease
(thromboangiitis obliterans), and, rarely, sys-
temic sclerosis (4, 40—42).

Renal disease is the most common feature of
classic polyarteritis (70% or greater) and is the
result of vasculitis, glomerulonephritis, or hyper-
tension (4). Aneurysms of the renal circulation
may rupture, leading to spontaneous perirenal
hemorrhage and hypotension (43). Recent data
indicate that even when end-stage renal disease
develops (often cited as the most common cause of
death in polyarteritis nodosa), the overall sur-
vival is quite good and comparable to that of age-
matched control patients with non-diabetes-asso-
ciated renal disease (62% at 36 months).
Furthermore, 10% of patients have been reported
to recover significant renal function and discon-
tinue dialysis (44).

Neurologic involvement is common to all
forms of SNV. Peripheral neuropathy occurs in up
to two-thirds of patients, usually in the form of a
diffuse sensorimotor polyneuropathy; however,
mononeuritis and isolated cutaneous neuropa-
thies may develop (45). The most common sites
are the peroneal, sural, radial, ulnar, and median
nerves (45). The most frequently identified cra-
nial neuropathies are cranial nerves II, VII, and
VIII (46). Central nervous system disease (40%) is
manifested by diffuse (encephalopathy, seizure)
or focal (cerebrovascular accident) abnormalities.

Gastrointestinal disease in patients with
SNV includes ischemia and infarction of the
bowel, hepatobiliary tree, pancreas, and appendix
(47^9). Clinically, individuals initially have fe-
ver, abdominal pain, gastrointestinal bleeding
(upper, lower, or intra-abdominal), peritonitis,
and intrahepatic hemorrhage (47^9).

Pulmonary infiltrates, characteristic of
Churg-Strauss vasculitis and Wegener's granulo-
matosis, are typically not seen in classic poly-
arteritis but may be a manifestation of overlap
SNV and microscopic polyarteritis. Pleural effu-
sions should suggest infection until proven other-
wise. Cardiac disease is characterized by conges-
tive heart failure, fibrosis and pericarditis

(Churg-Strauss), myocardial infarction and ar-
rhythmias (classic polyarteritis nodosa), and cor-
onary artery aneurysms (Kawasaki disease) (46,
50, 51).

The cutaneous lesions of SNV are pleomor-
phic and not distinct for any individual syndrome.
These include palpable purpura, urticaria, ulcers,
livedo reticularis, and subcutaneous nodules (52).

Testicular vasculitis, although common at
autopsy (up to 86%), is symptomatic in only 2%>-
18% of patients (53). Isolated lesions of necrotiz-
ing vasculitis (without associated systemic dis-
ease) involving the epididymis or testis have been
rarely reported and may be asymptomatic or
mimic neoplastic disease (53, 54).

Diagnostic Studies

The laboratory approach to the diagnosis of
SNV has changed dramatically over the past few
years. The observation that antineutrophil cyto-
plasmic autoantibodies (ANCA) are both sensi-
tive and specific for certain subsets of necrotizing
vasculitis has been critical to the diagnosis and
management of patients with suspected SNV.
The ANCA represent a new class of autoantibod-
ies directed against myeloid lysosomal enzymes.
Using indirect immunofluorescence on ethanol-
fixed polymorphonuclear leukocytes, two pat-
terns have been identified (55). The classic pat-
tern is cytoplasmic (c-ANCA), which has recently
been demonstrated to be against a 29-kDa serine
protease (proteinase 3) located in azurophilic (pri-
mary) granules and to a lesser extent on the
plasma membrane of PMN and monocytes (56).
Antibodies with a perinuclear pattern (p-ANCA)
are against myeloperoxidase and elastase, also in
the primary granules (55, 57). p-ANCA is found
with the greatest frequency in patients with dis-
ease confined to the kidney (crescentic glomeru-
lonephritis), whereas c-ANCA positivity is great-
est in patients with granulomatous lung disease
(58).

The sensitivity of the ANCA is greatest in
Wegener's granulomatosis, where they have been
noted in a prevalence of 50%-96% (59, 60). The
sensitivity is as high as 100% during active dis-
ease, 70% for disease limited to the respiratory
tract, and 30% during remission (57). The speci-
ficity for Wegener's is also very high (90% or
greater) (60).

c-ANCA has also been noted in patients with
microscopic polyarteritis, a necrotizing vasculitis
involving the kidneys with pathology that resem-
bles that seen with Wegener's (61), and Kawasaki

Vol. 60 No. 2

NECROTIZING VASCULITIS— KAUFMAN & KAPLAN

107

disease. Antimyeloperoxidase antibodies have
been identified in idiopathic crescentic glomeru-
lonephritis, Churg-Strauss, Takayasu vasculitis,
and less often in systemic lupus erythematosus,
relapsing polychondritis, Behcet's disease, He-
noch-Schonlein purpura (IgA isotype), inflamma-
tory bowel disease, and primary sclerosing chol-
angitis (57, 62-64). In addition to their central
role in the evaluation of patients with SNV, cur-
rent evidence supports a pathogenic function for
ANCA. The ANCA-specific antigens are translo-
cated to the surface of cells following exposure to
cytokines such as TNF-a. Subsequent binding of
ANCA to these primed PMN has demonstrated
release of oxygen radicals and proteases in vitro
(65). The recent finding that ANCA may be pro-
duced in the respiratory tract of patients with
Wegener's provides additional support for their
potential pathogenetic role (66).

Antineutrophil autoantibodies directed
against the 29-kDa serine protease and myelo-
peroxidase are predominantly of the IgGl and
IgG4 isotypes (67). IgG4 is produced after recur-
rent stimulation and is dependent on T-cell pro-
duction of 11-4, which regulates the isotype switch
from IgGl to IgG4 (68). It is therefore likely that,
as a result of an antigen-driven T-cell response in
Wegener's, 11-4 is generated and is important to
the synthesis of ANCA, which subsequently acti-
vate PMN and contribute to tissue injury.

Recent studies have suggested that the pro-
myelocjd;ic cell line HL60 is useful for differenti-
ating antineutrophil antibodies from antinuclear
antibodies (ANA) (69). These cells are potentially
helpful in distinguishing a false positive p-ANCA
from an ANA; however, because of phenotypic
drift during cell passage, ANCA-specific antigens
may be lost (70).

In addition to the ANCA, other autoantibod-
ies have been reported to be present in patients
with SNV. None, unfortunately, have the same
degree of sensitivity or specificity, but may offer
insight into the mechanisms of tissue injury. An-
tiendothelial cell autoantibodies, identified pre-
dominantly in patients with small vessel vasculi-
tis, have been reported in Kawasaki disease (71).
The antiphospholipid antibodies are primarily as-
sociated with thromboembolic disease and a non-
inflammatory vasculopathy rather than a true
vasculitis; however, there is some evidence for the
latter as well (72, 73). Antilamin antibodies are a
rare group of autoantibodies that have been found
in association with small vessel vasculitis (74).
The Ro (SS-A) antigen, associated with primary
Sjogren's syndrome and systemic lupus, has been

linked to cutaneous vasculitis in Sjogren's (75).
Anti-IgE autoantibodies are of theoretical value
in understanding the mechanisms of histamine
release from basophils and mast cells and have
been identified in individuals with vascular le-
sions such as systemic lupus and scleroderma (26,
76), as well as in some patients with urticarial
vasculitis (77). Unidentified precipitins against
rabbit thymus extract not found in other connec-
tive tissue diseases or polyarteritis nodosa have
been identified in rheumatoid vasculitis (78).

Visceral angiography is useful in the evalu-
ation of SNV and demonstrates microaneurysms
in 30%-60% of patients (79). These are most often
present in the renal circulation, the celiac axis,
and the superior mesenteric artery, necessitating
study of each of these vascular beds during arte-
riography (4, 46). Unfortunately, the finding of
microaneurysms is nonspecific, and they may be a
feature of disorders that mimic SNV, such as
atrial myxoma, endocarditis, fibromuscular dys-
plasia, and pseudoxanthoma elasticum (4). Fur-
thermore, caution should be exercised in perform-
ing angiography in view of reports of dye-
associated acute renal failure in the setting of
SNV (80).

Ultimately, the most conclusive diagnosis is
dependent on demonstrating the lesion of SNV
histologically. In general, biopsy sites should be
selected as determined by the clinical manifesta-
tions, and tissue should be obtained from the most
easily accessible and involved organ. Cutaneous
lesions will often demonstrate vasculitis; how-
ever, this does not necessarily correlate with the
presence of visceral involvement. Nevertheless,
in an individual with multisystem disease and
vasculitic skin lesions, additional tissue need not
be obtained. Abnormal neurophysiologic studies
correlate well with the ability to define vasculitis
by sural nerve biopsy (81). Although the yield
from muscle biopsy is traditionally greatest when
painful areas are sampled, blind biopsies have
also been reported to be useful (82). Recent re-
ports of rectal biopsies demonstrating necrotizing
vasculitis may suggest an alternative "blind" site
in patients with diffuse systemic disease in whom
a SNV is suspected clinically (83). Renal biopsies
most often demonstrate glomerulonephritis in pa-
tients with SNV, and microaneurysms are a po-
tential source of intrarenal hemorrhage; how-
ever, percutaneous needle biopsy by an
experienced individual may also be useful (84). In
Wegener's granulomatosis, recent studies have
demonstrated diagnostic nasal biopsies in as
many as 53% of patients, emphasizing the need

108

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

Confirming a Diagnosis of Vasculitis

Histology (biopsy site) Angiography
Skin Renal arteries

Sural nerve Celiac axis

Muscle Superior
I mesenteric artery

Rectal

Renal

Fig. Proposed diagnostic approach to biopsy and an-
giography in systemic vasculitis. The choice of a blind
biopsy or visceral angiography will often depend on the
clinical presentation and risk-benefit analysis for a par-
ticular patient.

for obtaining tissue specimens larger than 5 mm
(85). A suggested approach to biopsy is outlined in
the Fig.

Differential Diagnosis
and Therapy

A number of clinical syndromes and diseases
may mimic SNV and must be differentiated from
it. In that regard, the recent criteria established
by the American College of Rheumatology (86)
will be helpful in distinguishing SNV from clini-
cally similar entities and allow standardization of
patients entered into prospective clinical trials.
Embolic disorders, in particular, may masquer-
ade as vasculitis. The syndrome of cholesterol or
atheroemboli presents with vasculitic-appearing
skin lesions, livedo reticularis, renal insuffi-
ciency, and eosinophilia (87). Similar cutaneous
and renal findings may occur in the syndrome of
left atrial myxoma (88). Thrombotic thrombocy-
topenic purpura is a multisystem syndrome char-
acterized by fever, renal disease, microangio-
pathic anemia, thrombocytopenia, and central
nervous system abnormalities that has been re-
ported to be associated with microaneurysms (4).
Finally, the anticardiolipin or antiphospholipid
syndrome is a thromboembolic disorder associ-
ated with cutaneous and visceral infarction.

Corticosteroid and immunosuppressive
agents provide the principal therapy in these dis-
eases. In an open trial with cyclophosphamide (2
mg/kg per day) for severe necrotizing systemic
vasculitis, dramatic results were reported in 16
patients with progressive disease in spite of high

doses of corticosteroids (89). Although the use of
cytotoxic agents has had a dramatic impact on the
course of SNV compared to historical controls, it
should be emphasized that there have been very
few attempts at prospective controlled studies.
Furthermore, acute leukemia following cyclo-
phosphamide therapy for PAN stands as a re-
minder of the significant toxicity of alkylating
drugs (90). A recent multicenter, prospective,
randomized study examined the risks and bene-
fits of corticosteroids and plasma exchange with
and without cyclophosphamide for PAN and
Churg-Strauss vasculitis in 71 patients (91). In
this trial there were more withdrawals due to
lack of efficacy in the group without cyclophos-
phamide, and more related to toxicity in the cy-
totoxic group. Although the ten-year cumulative
survival was no different in either group (72%
versus 75%), relapse occurred statistically less of-
ten in the patients receiving cyclophosphamide.
Newer, and more novel, immunomodulatory ap-
proaches to therapy have included monoclonal
anti-CD4 antibodies (92) and high-dose intrave-
nous immunoglobulin (93).

The use of corticosteroids for the treatment of
SNV has recently been challenged by Conn et al.
(94). This hypothesis suggests that although ste-
roids may blunt much of the acute inflammatory
process, platelet thromboxane may not be inhib-
ited in vivo. As a result, platelet activation may
lead to intimal proliferation and luminal occlu-
sion on the basis of a noninflammatory vasculop-
athy.

At present, no conclusive data exist regard-
ing the benefit of different immunosuppressive
agents alone or in combination with other modal-
ities. Indeed, as others have noted, a large multi-
center study utilizing recently established crite-
ria for PAN and related disorders will be required
to help define optimal therapy for SNV (95).

References

1. Christian CL. Vasculitis: genus and species (editorial).

Ann Intern Med 1984; 101:862-863.

2. Kaufman LD. Polyarteritis. In: Taylor RB, ed. Difficult

diagnosis II. Philadelphia: WB Saunders, 1991 (in
press).

3. Alarcon-Segovia D. Classification of the necrotizing vas-

culitides in man. Clin Rheum Dis 1980; 6:223-231.

4. Cupps TR, Fauci AS. The vasculitides. In: Smith LJ, ed.

Major problems in internal medicine, vol. 21. Philadel-
phia: WB Saunders, 1981.

5. Kussmaul A, Maier K. Uber eine bischer nicht be-

schreibene eigenthumliche arterienerkrankung (peri-
arteritis nodosa), die mit morbus brightii und rapid
fortschreitender allgemeiner muskellahmung ein-
hergeht. Dtsch Arch Klin Med 1866; 1:484-517.

6. Fauci AS, Haynes BF, Katz P. The spectrum of vasculi-

Vol. 60 No. 2

NECROTIZING VASCULITIS— KAUFMAN & KAPLAN

109

tis — clinical, pathologic, immunologic, and therapeutic
considerations. Ann Intern Med 1978; 89(Part 1):660-
676.

7. Zeek PM, Smith CC, Weeter JC. Studies on periarteritis

nodosa. III. The differentiation between the vascular
lesions of periarteritis nodosa and of hypersensitivity.
Am J Pathol 1948; 24:889-917.

8. Lie JT. Diagnostic histopathology of major systemic and

pulmonary vasculitic syndromes. Rheum Dis Clin
North Am 1990; 16:269-292.

9. Cochrane CG, Koffler D. Immune complex disease in ex-

perimental animals and man. Adv Immunol 1973; 16:
185-264.

10. Bielory L, Gascon P, Lawley T, Young NS, Frank MM.

Human serum sickness: a prospective analysis of 35
patients treated with equine anti-thymocyte globulin
for bone marrow failure. Medicine 1988; 67:40—57.

11. Schrieber L, Penny R. Factors influencing immune com-

plex localisation. Rheumatol Int 1984; 4:95-109.

12. Christian CL. Hepatitis B virus (HBV) and systemic vas-

culitis (editorial). Clin Exp Rheumatol 1991; 9:1-2.

13. Marcellin P, Calmus Y, Takahashi H, Zignego AL,

Chatenoud L, Galanaud LP, Leibowitch M, Bach JF,
Benhamou JP, Tiollais P, Wands J, Guillevin L,
Brechot C. Latent hepatitis B virus (HBV) infection in
systemic necrotizing vasculitis. Clin Exp Rheumatol
1991; 9:23-28.

14. Doherty M, Bradfield JWB. Polyarteritis nodosa associ-

ated with acute cytomegalovirus infection. Ann Rheum
Dis 1981; 40:419-421.

15. Li Loong TC, Coyle PV, Anderson MJ, Allen GE, Connolly

JH. Human serum parvovirus associated vasculitis.
Postgrad Med J 1986; 62:493-494.

16. Calabrese LH, Estes M, Yen-Lieberman B, Proffitt MR,

Tubbs R, Fishleder AJ, Levin KH. Systemic vasculitis
in association with human immunodeficiency virus in-
fection. Arthritis Rheum 1989; 32:569-576.

17. Linnemann CC, Alvira M. Pathogenesis of the varicella-

zoster angiitis in the CNS. Arch Neurol 1980; 37:239.

18. Hoffman GS, Franck WA. Infectious mononucleosis au-

toimmunity, and vasculitis. JAMA 1979; 241:2735-
2736.

19. Larsson A, Forsgren M, Hard AF, Segerstad H, Strander

H, Cantrell K. Administration of interferon to an infant
with congenital rubella syndrome involving persistent
viremia and cutaneous vasculitis. Acta Pediatr Scand
1976; 65:105-110.

20. Lang GE, Schonherr U, Naumann GOH. Retinal vasculi-

tis with proliferative retinopathy in a patient with ev-
idence of Borrelia burgdorferi. Am J Ophthalmol 1991;
111:243-244.

21. Ferri C, Greco F, Longombardo G, Palla P, Marzo E, Mo-

retti A. Hepatitis C virus antibodies in mixed cryoglob-
ulinemia. Clin Exp Rheumatol 1991; 9:95-96.

22. Sergent JS, Christian CL. Necrotizing vasculitis after

acute serous otitis media. Ann Intern Med 1974; 56:
412-416.

23. Komadina KH, Houk RW. Periarteritis nodosa presenting

as recurrent pneumonia following splenectomy for
hairy cell leukemia. Semin Arthritis Rheum 1989; 18:
252-257.

24. Kulp-Shorten CL, Rhodes RH, Peterson H, Callen JP. Cu-

taneous vasculitis associated with pheochromocytoma.
Arthritis Rheum 1990; 33:1852-1856.

25. Baeza ML, Reddigari S, Haak-Frendscho M, Kaplan AP.

Purification and further characterization of human

mononuclear cell histamine-releasing factor. J Clin In-
vest 1989; 83:1204-1210.

26. Kaufman LD, Gruber BL, Marchese M, Seibold J. Anti-

IgE autoantibodies in systemic sclerosis (scleroderma).
Ann Rheum Dis 1989; 48:201-205.

27. Braverman IM, Yen A. Demonstration of immune com-

plexes in spontaneous and histamine-induced lesions
and in normal skin of patients with leukocytoclastic
angiitis. J Invest Dermatol 1975; 64:105-112.

28. Baggiolini M, Walz A, Kunkel S. Neutrophil activating

peptide- 1/interleukin 8, a novel cytokine that activates
neutrophils. J Clin Invest 1989; 84:1045-1049.

29. Walz A, Dewald B, von Tscharner V, Baggiolini M. Effects

of the neutrophil-activating peptide NAP-2, platelet ba-
sic protein, connective tissue-activating peptide III, and
platelet factor 4 on human neutrophils. J Exp Med
1989; 170:1745-1750.

30. Pober JS. Cytokine-mediated activation of vascular endo-

thelium: physiology and pathology. Am J Pathol 1988;
133:426-433.

31. Anon. Granulomas and cytokines. Lancet 1991; 1:1067-

1068.

32. Tai P, Holt ME, Denny P, Gibbs AR, Williams BD, Spry

CJF. Deposition of eosinophil cationic protein in gran-
ulomas in allergic granulomatosis and vasculitis: the
Churg-Strauss syndrome. Br Med J 1984; 289:400-402.

33. Farmer KL, Kurzrock R, Gutterman JU, Duvic M. Nec-

rotizing vasculitis at granulocyte-macrophage-colony-
stimulating factor injection sites. Arch Dermatol 1990;
126:1243-1244.

34. Proud D, Kaplan AP. Kinin formation: mechanisms and

role in inflammatory disorders. Rev Immunol 1988; 6:
49-83.

35. Bevilacqua MP, Schleef RR, Gimbrone MA, Loskutoff DJ.

Regulation of the fibrinolytic system of cultured human
vascular endothelium by interleukin-1. J Clin Invest
1986; 78:587-591.

36. Nawroth PP, Stern DM. Modulation of endothelial cell

hemostatic properties by tumor necrosis factor. J Exp
Med 1986; 163:740-745.

37. Churg J, Strauss L. Allergic granulomatosis, allergic an-

giitis, and periarteritis nodosa. Am J Pathol 1951; 27:
277-301.

38. Kaufman LD, Kaplan AP. Microscopic polyarteritis. Hosp

Pract 1989; 24(June):85-104.

39. Gilliam JH, Challa VR, Agudelo CA. Vasculitis involving

muscle associated with Crohn's colitis. Gastroenterol-
ogy 1981; 81:787-790.

40. Scott DGI, Bacon PA, Tribe CR. Systemic rheumatoid vas-

culitis: a clinical and laboratory study of 50 patients.
Medicine 1981; 60:288-297.

41. Cogan's syndrome. Lancet 1991; 1:1011-1012.

42. Pathak R, Gabor AJ. Scleroderma and central nervous

system vasculitis. Stroke 1991; 22:410-413.

43. Smith DL, Wernick R. Spontaneous rupture of a renal

artery aneurysm in polyarteritis nodosa: critical review
of the literature and report of a case. Am J Med 1989;
87:464-467.

44. Nissenson AR, Port FK. Outcome of end-stage renal dis-

ease in patients with rare causes of renal failure. III.
Systemic/vascular disorders. Q J Med 1990; 74:63-74.

45. Moore P, Fauci A. Neurologic manifestations of systemic

vasculitis: a retrospective and prospective study of the
clinicopathologic features and responses to therapy in
25 patients. Am J Med 1981; 71:517-524.

46. Vertzman L. Polyarteritis nodosa. Clin Rheum Dis 1980;

6:297-317.

110

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

47. Camilleri M, Pusey CD, Chadwick VS, Rees AJ. Gastro-

intestinal manifestations of systemic vasculitis. Q J
Med 1983; 52:141-149.

48. Zizic TM, Classen JN, Stevens MB. Acute abdominal com-

plications of systemic lupus erythematosus and poly-
arteritis nodosa. Am J Med 1982; 73:525-531.

49. AUeman M, Janssens A, Spoelstra P, Kroon H. Spontane-

ous intrahepatic hemorrhages in polyarteritis nodosa.
Ann Intern Med 1986; 105:712-713.

50. Cupps TR, Springer RM, Fauci AS. Chronic, recurrent

small-vessel cutaneous vasculitis: clinical experience in
13 patients. JAMA 1982; 247:1994-1998.

51. Hasley PB, Follansbee WP, Coulehan JL. Cardiac mani-

festations of Churg-Strauss syndrome: report of a case
and review of the literature. Am Heart J 1990; 4:996-
999.

52. Kaufman LD. The skin and vasculitis. In: LeRoy EC, ed.

Systemic vasculitis. Marcel Dekker, 1991 (in press).

53. Shurbaji MS, Epstein JI. Testicular vasculitis: implica-

tions for systemic disease. Hum Pathol 1988; 19:186-
189.

54. Huisman TK, Collins WT Jr, Voulgarakis OR. Polyarteri-

tis nodosa masquerading as a primary testicular neo-
plasm: a case report and review of the literature. J Urol
1990; 144:1236-1238.

55. Kallenberg COM, Cohen Tervaert JW, van der Woude FJ,

Goldschmeding R, von dem Borne AEGKr, Weening JJ.
Autoimmunity to lysosomal enzymes: new clues to vas-
culitis and glomerulonephritis? Immunol Today 1991.

56. Csernok E, Ludemann J, Gross WL, Sainton DF. Ultra-

structural localization of proteinase 3, the target anti-
gen of anti-cytoplasmic antibodies circulating in We-
gener's granulomatosis. Am J Pathol 1990; 137:1113-
1120.

57. Ramirez G, Khamashta MA, Hughes GRV. The ANCA

test: its clinical relevance. Ann Rheum Dis 1990; 49:
741-742.

58. Falk RJ, Jennette JC. Wegener's granulomatosus, sys-

temic vasculitis, and antineutrophil cytoplasmic au-
toantibodies. Annu Rev Med 1991; 42:459^69.

59. Savage COS, Winearls CG, Jones S, Marshall PD, Lock-

wood CM. Prospective study of radioimmunoassay for
antibodies against neutrophil cytoplasm in diagnosis of
systemic vasculitis. Lancet 1987; 1:1389—1393.

60. Nolle B, Specks U, Ludemann J, Rohrbach MS, DeRemee

RA, Gross WL. Anticytoplasmic autoantibodies: their
immunodiagnostic value in Wegener granulomatosis.
Ann Intern Med 1989; lll:28--i0.

61. Savage COS, Winearls CG, Evans DJ, Rees AJ, Lockwood

CM. Microscopic polyarteritis: presentation, pathology
and prognosis. Q J Med 1985; 56:467-483.

62. Cohen Tervaert JW, Goldschmeding R, Elema JD, von

dem Borne AEGK, Kallenberg CGM. Antimyeloperox-
idase antibodies in Churg-Strauss syndrome. Thorax
1991; 46:70-71.

63. Falk RJ. ANCA-associated renal disease. Kidney Int

1990; 38:998-1010.

64. Duerr RH, Targan SR, Landers CJ, LaRusso NF, Lindsay

KL, Wiesner RH, Shanahan F. Neutrophil cytoplasmic
antibodies: a link between primary sclerosing cholan-
gitis and ulcerative colitis. Gastroenterology 1991; 100:
1385-1391.

65. Falk RJ, Terrell RS, Charles LA, Jennette JC. Anti-neu-

trophil cytoplasmic autoantibodies induce neutrophils
to degranulate and produce oxygen radicals in vitro.
Proc Natl Acad Sci USA 1990; 87:4115-^119.

66. Baltaro RJ, Hoffman GS, Sechler JMG, Suffredini AF,

Shelhamer JH, Fauci AS, Fleisher TA. Immunoglobu-
lin G antineutrophil cytoplasmic antibodies are pro-
duced in the respiratory tract of patients with Wegen-
er's granulomatosus. Am Rev Respir Dis 1991; 143:
275-278.

67. Brouwer E, Cohen Tervaert JW, Horst G, Huitema MG,

Van Der Giessen M, Limburg PC, Kallenberg CGM.
Predominance of IgGl and IgG4 subclasses of anti-neu-
trophil cytoplasmic autoantibodies (ANCA) in patients
with Wegener's granulomatosus and clinically related
disorders. Clin Exp Immunol 1991; 83:379-386.

68. Spiegelberg HL. Biological role of different antibody

classes. Int Arch Allergy AppI Immunol 1989; 90:22-
27.

69. Christenson VD, Dooley MA, Allen NB. Discrimination ol

antineutrophil antibodies from antinuclear antibodies
using immunofluorescence on neutrophils and HL6C
cells. J Rheumatol 1991; 18:575-579.

70. Charles LA, Jennette JC, Falk RJ. The role of HL60 cells

in the detection of antineutrophil cytoplasmic autoanti-
bodies (editorial). J Rheumatol 1991; 18:491^94.

71. Baguley E, Hughes GRV. Antiendothelial cell antibodies,

J Rheumatol 1989; 16:716-717.

72. Alarcon-Segovia D, Cardiel MH, Reyes E. Antiphospho-

lipid arterial vasculopathy. J Rheumatol 1989; 16:762-
767.

73. Lie JT. Vasculopathy in the antiphospholipid syndrome:

thrombosis or vasculitis, or both? J Rheumatol 1989:
16:713-715.

74. Lassoued K, Guilly MN, Danon F, Andre C, Dhumeaux D,

Clauvel JP, Brouet JC, Seligmann M, Courvalin JC,
Antinuclear autoantibodies specific for lamins: charac-
terization and clinical significance. Ann Intern Med
1988; 108:829-833.

75. Alexander EL, Provost TT. Cutaneous manifestations oi

primary Sjogren's syndrome: a reflection of vasculitis
and association with anti-Ro (SS-A) antibodies. J Invest
Dermatol 1983; 80:386-391.

76. Gruber BL, Kaufman LD, Marchese M, Kaplan AP. Anti-

IgE autoantibodies in systemic lupus erythematosus:
prevalence and functional significance. Arthritis
Rheum 1988; 31:1000-1006.

77. Gruber BL, Baeza ML, Marchese MJ, Agnello V, Kaplan

AP. Prevalence and functional role of anti-IgE au-
toantibodies in urticarial syndromes. J Invest Dermatol
1988; 90:213-217.

78. Scott GI, Skinner RP, Bacon PA, Maddison PJ. Precipi-

tating antibodies to nuclear antigens in systemic vas-
culitis. Clin Exp Immunol 1984; 56:601-606.

79. Stanson AW. Roentgenographic findings in major vascu-

litic syndromes. Rheimi Dis Clin North Am 1990; 16:
293-308.

80. Kaur JS, Goldberg JP, Schrier RW. Acute renal failure

following arteriography in a patient with polyarteritis
nodosa. JAMA 1982; 247:833-834.

81. Wees SJ, Sunwoo IN, Oh SJ. Sural nerve biopsy in sys-

temic necrotizing vasculitis. Am J Med 1981; 71:525-
532.

82. Dahlberg PJ, Lockhart JM, Overholt EL. Diagnostic stud-

ies for systemic necrotizing vasculitis: sensitivity, spec-
ificity, and predictive value in patients with multisys-
tem disease. Arch Intern Med 1989; 149:161-165.

83. Tribe CR, Scott DGI, Bacon P. Rectal biopsy in the diag-

nosis of systemic vasculitis. J Clin Pathol 1981; 34:843-
850.

84. Scott DGI, Bacon PA, Elliott PJ, Tribe CR, Wallington

TB. Systemic vasculitis in a district general hospital

Vol. 60 No. 2

NECROTIZING VASCULITIS— KAUFMAN & KAPLAN

111

1972-1980: clinical and laboratory features, classifica-
tion and prognosis of 80 cases. Q J Med 1982; 203:292-
311.

85. Del Buono EA, Flint A. Diagnostic usefulness of nasal

biopsy in Wegener's granulomatosis. Hum Pathol 1991;
22:107-110.

86. Lightfoot RWJ, Michel BA, Bloch DA, Hunder GO, Zvai-

fler MJ, McShane DJ, Arend WP, Calabrese LH,
Leavitt R, Lie JT, Masi AT, Mills J, Stevens M, Wallace
S. The American College of Rheumatology 1990 criteria

I for the classification of polyarteritis nodosa. Arthritis

I Rheum 1990; 33:1088-1093.

87. Cappiello RA, Espinoza LR, Adelman H, Aguilar J, Vasey

F, Germain BF. Cholesterol embolism: a pseudovascu-
I litic syndrome. Semin Arthritis Rheum 1989; 18:240-
246.

88. Feldman AR, Keeling JH. Cutaneous manifestation of

atrial myxoma. J Am Acad Dermatol 1989; 21:1080-
1084.

89. Fauci AS, Katz P, Haynes BF, Wolff SM. Cyclophospha-

mide therapy of severe systemic necrotizing vasculitis.
N Engl J Med 1979; 301:235-238.

90. Escalante A, Kaufman RL, Beardmore TD. Acute myelo-

cytic leukemia after the use of cyclophosphamide in the

treatment of polyarteritis nodosa. J Rheumatol 1989;
16:1147-1149.

91. Guillevin L, Jarrousse B, Lok C, Lhote F, Jais JP, Le Thi

Huong Du D, Bussel A, and the Cooperative Study
Group for Polyarteritis Nodosa. Longterm follow-up af-
ter treatment of polyarteritis nodosa and Churg-
Strauss angiitis with comparison of steroids, plasma ex-
change and cyclophosphamide to steroids and plasma
exchange. A prospective randomized trial of 71 pa-
tients. J Rheumatol 1991; 18:567-574.

92. Mathieson PW, Cobbold SP, Hale G, Clark MR, Oliveira

DBG, Lockwood CM, Waldmann H. Monoclonal anti-
body therapy in systemic vasculitis. N Engl J Med
1990; 323:250-254.

93. Jayne DRW, Davies MJ, Fox CJV, Black CM, Lockwood

CM. Treatment of systemic vasculitis with pooled in-
travenous immunoglobulin. Lancet 1991; 337:1137-
1139.

94. Conn DL, Tompkins RB, Nichols WL. Glucocorticoids in

the management of vasculitis — a double edged sword?
(editorial). J Rheumatol 1988; 15:1181-1183.

95. Conn DL. Role of cyclophosphamide in treatment of poly-

arteritis nodosa? (editorial). J Rheumatol 1991; 18:489-
490.

Grand
Rounds

The Global Impact of Penicillin:

Then and Now

Gene H. Stollerman, M.D.

Nineteen ninety-one was the 50th anniversary
of the introduction of penicillin into medical prac-
tice, thanks largely to Ernst Chain's gift for mas-
terminding its industrial production (1).

In the Beginning. In 1941, those of us in
medical school were soon to become appalled and
enthralled by the great scourges of young hearts,
rheumatic fever, bacterial endocarditis, and syph-
ilis. Only for the last of these scourges was treat-
ment available, and that consisted only of arseni-
cals taken for prolonged periods.

In 1944, at the peak of World War II, when I
began house staff training at the Mount Sinai
Hospital in New York, penicillin was a precious
military priority. Small supplies were made
available to leading authorities for clinical inves-
tigation. Among these were George Baehr, Chief
of Medicine, who shared with Emanuel Libman
and other Sinai greats an international reputa-
tion in the study of bacterial endocarditis. Baehr
and Gregory Shwartzman set up a laboratory un-
der Stanley Schneierson that was among the ear-
liest to assay the sensitivity of bacteria to peni-
cillin tissue levels following various treatment
regimens. We, the house staff, were the labor
crew for the study of the earliest treatment of
subacute bacterial endocarditis (SBE) (2).

Adapted from the author's Stanley Seckler Memorial Lecture
at Mount Sinai School of Medicine, New York, December 10,
1991. Final manuscript received June 1992.

From the ENRM Veterans Affairs Hospital, Bedford, MA,
and the Dept. of Medicine and School of Public Health, Boston
Univ. School of Medicine. Address reprint requests to the au-
thor. Associate Chief of Staff for Geriatrics and Extended Care
and Director/HSR&D Field Program, ENRM Veterans Hospi-
tal (180), 200 Springs Road, Bedford, MA 01730.

Repository Penicillins

When I returned from military service ii
1947 to become George Baehr's chief resident, re
search on penicillin treatment of SBE was in ful
swing. Ed Roston, my co-chief resident, and
were given the job of testing the newly appearin;
repository penicillins, specifically procaine peni
cillin G.

Levels and Duration of Therapy. From th
studies of Harry Eagle (3) (Fig. 1), it was appar
ent that low levels of penicillin were all that wer
needed to kill sensitive organisms such as th
group A Streptococcus (GrAS). The time require^
to kill virtually all organisms in a cultur
reached a maximum in but a few hours at a per
icillin level as low as 0.01 U/mL. No greater cor
centration of penicillin could increase the rate c
killing when the organism was in its maxima
growth phase. For a slow-growing organism c
similar sensitivity, such as Treponema pallidurri
the maximum rate of killing also occurred at o
below a level of 0.01 u/mL, but the time requirei
for killing an organism that multiplied every sev
eral hours instead of every 20 minutes was mucl
greater.

It was clear, then, that long duration of ther
apy at low levels was what was needed. Th
emerging repository penicillins could provide jus
those prolonged low levels without continuous in
travenous therapy or frequent bolus injections.

Another critical principle of successful peni
cillin therapy was established by the studies o
Dr. Barry Wood (4) (Fig. 2). When penicillin wa
added to a growing culture of S. pneumoniae dur
ing the log phase of its growth, the organism
were rapidly killed. If, however, penicillin wa
added when the culture stopped growing, the or
ganisms were not killed. When pneumococci wer

112

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 199

Vol. 60 No. 2

PENICILLIN— STOLLERMAN

113

inoculated into pus which prevented their
growth, peniciUin did not kill until the infected
pus was reinoculated into fresh culture medium.

The lesson was clear for the treatment of bac-
terial endocarditis. Given a sensitive organism in
the active phase of growth, a modest level of pen-
icillin should suffice, but the duration of therapy
would have to exceed the time required for an
abscess in a valve to clear itself of dormant
though potentially still viable organisms.

SBE is an infection in which host defenses
are virtually ineffectual. The outcome is a contest
between antibiotic and parasite. An in vitro test
became our experimental model to determine at
what level of antibiotic we could expect a bacte-
ricidal rather than a bacteriostatic effect. It al-
ready had been determined that bacteriostatic
antibiotics would not cure SBE. Some organisms,
in fact, could be efficiently eliminated only by a
two-pronged antibiotic attack, one that became
available with the introduction of streptomycin
and other aminoglycosides.

This in vitro determination of bacteriostatic
vs. bacteriocidal actions of antibiotics became
known as the "Schlichter" test (5). The minimum
levels required to kill an organism became known
as the minimum bactericidal concentration
(MBC), and the minimum levels required to in-
hibit growth, the minimum bacteriostatic con-

centration (MIC). When an organism was killed
by penicillin at the same level as it was inhibited
(MBC/MIC = 1) it was judged to be a strain
whose destruction was close to assured in SBE
therapy. When, however, a marked difference oc-
curred between MIC and MBC, one could antici-
pate a high risk of relapse unless combined ther-
apy with two bacteriocidal antibiotics eliminated
the disparity. The latter situation was encoun-
tered most often in the treatment of bacterial en-
docarditis due to enterococci and staphylococci,
but rarely with the viridans streptococci and
never with group A streptococci or pneumococci.

Armed with these principles, Ed Roston and I
(5) conducted a clinical experiment (Table 1) us-
ing relatively small doses of procaine penicillin G
for the treatment of SBE due to sensitive strains
of Streptococcus viridans. To effect a cure we
needed only to sustain blood levels of approxi-
mately 1.0 u/mL for 6-7 weeks by giving procaine
penicillin twice daily in doses ranging from
450,000 to 600,000 units (6).

Cellular Action of Penicillin

By the 1950s, the mode of action of penicillin
at the cellular level had become clearer. Interfer-
ence with cell-wall synthesis could be readily
demonstrated in some strains of E. coli. When

noor

TIME M HOURS AT 37*C

Fig. I. Rate of killing of growing cultures of group A streptococci at varying
concentrations of penicillin G (Ref. 3).

114

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

PENICILLIN

ADDED TO C

A - GROWTH CURVE OF PNEUMOCOCCIS I IN BROTH
B • PENICILLIN ADDED IN LOGARITHMIC PHASE
C - PENICILLIN ADDED IN STATIONARY PHASE

8 12 16

INCUBATION (HOURS)

Fig. 2. Rapid killing of S. pneumoniae by penicillin G when
added to culture during log phase of growth, in contrast to lack
of significant killing when added during stationary phase of
growth (Ref. 4).

grown in penicillin, synthesis of the cell wall was
inhibited, but the organism's inner membrane-
bound protoplast could survive with variable ro-
bustness against osmotic lysis, especially when
cultured in media of high osmotic concentration.
The cellular metabolic basis for penicillin's ac-
tion, meanwhile, was rapidly elucidated by the
discovery of the inhibition by penicillin of bacte-
rial cell wall synthesis (7). Cross-linking of the
muramic acid-N-acetyl glucosamine polysaccha-
ride chains of the cell wall by glycopeptides was
shown to be disrupted by penicillin with conse-
quent dissolution of the cell wall.

Penicillin Resistance. Further study has re-
vealed that gram-positive organisms have in
their cell membranes certain proteins that are
important in the regulation of normal cell-wall
lysis and synthesis and are necessary for the
growth, division, and remodeling of the bacterial
wall. The affinity of penicillin for these so-called
penicillin-binding-proteins (PBP) is essential for
penicillin's antibiotic effect (8). Organisms very
sensitive to penicillin lyse and explode. The omi-
nous defense that certain bacteria can develop by
producing beta lactamases that destroy penicillin
have made treatment of certain strains of staph-
ylococci, among others, notoriously difficult, re-
quiring, first, the development of methicillin and
its congeners for protection of the beta lactam
ring of penicillin against beta lactamases, and
later, alternatively, the introduction of the ceph-
alosporins. Methicillin resistance, however, has
now emerged in strains that have mutated to pro-
duce in their cell membranes PBPs of frighten-
ingly low penicillin affinity. They are thus "met-

abolically resistant," in contrast to resistance
resulting from beta lactamase production.

The cell walls of gram-negative organisms
have a more complex structure. An outer cell
membrane makes penetration of the cell wall by
penicillin possible only through surface porins, or
channels. Numerous genetically polymorphic
PBPs with varying affinity for penicillin are ei-
ther present or are inducible, so that resistance to
penicillin can rapidly emerge by numerous mech-
anisms.

Bacterial Adherence. Fortunately, with a
few rare exceptions, gram-negative organisms do
not stick to heart valves or endocardium in the
course of bacteremia. Adherence of streptococci
has been shown by my former colleagues, the late
Edwin Beachey and Itzhak Ofek (9), to be due to
surface lipoteichoic acids. GrAS adhering to a
buccal mucosal cell (Fig. 3) was shown to be due
not to the surface M protein, as we originally
thought, but to the closely associated membrane
lipid, lipoteichoic acid. Lipoteichoic acid is inter-
woven with streptococcal M protein, but its fatty
acid terminal protrudes from the cell membrane
and forms the ligand that binds to fibronectin on
the surface of most mucus membranes, and prob-
ably on scarred heart valves as well (10). Binding
by various lipoteichoic acids occurs with the
gram-positive organisms that cause SBE: strepto-
cocci, pneumococci, staphylococci, and entero-
cocci. The treatment of SBE due to methicillin-
resistant staphylococci, especially adherent to
prosthetic valves, has now become something of a
therapeutic nightmare.

Penicillin Therapy of
Sensitive Organisms

Rheumatic Fever. A more pleasant topic is
the impact of penicillin on group A streptococci

TABLE 1

Early Success in Procaine Penicillin RX of Subacute
Bacterial Endocarditis due to S. viridans

Penicillin
levels

Pen.

Days

Daily

Case

sens.

dose IM

Mean

Range

M.K.

0.1

450,000

1.5

0.2-2.0

BID

A.K.

0.14

450,000

1.1

0.5-1.3

BID

L.C.

0.2

600,000

1.45

0.6-2.0

BID

S.N.

0.05

450,000

1.0

0.8-1.3

BID

Adapted from ref. 6.

Vol. 60 No. 2

PENICILLIN— STOLLERMAN

115

and on T. pallidum. In 1950, the Wyeth Co. de-
veloped benzathine penicillin G (Bicillin). At the
time, I was director of Irvington House, a re-
search and treatment center for children with
rheumatic fever and rheumatic heart disease at
Irvington-on-Hudson, NY. Because of my interest
in long-acting penicillins and the prevention of
rheumatic fever, I was contacted to study Bicil-
lin's potential for human use. To our astonish-
ment, blood levels following a single intramuscu-
lar dose were shown to outlast, by far, all other
forms of repository penicillin salts. In fact, one
dose of 600,000 units, as Jerome Rusoff and I first
reported in JAMA in 1952 (11), persisted in blood
samples for 10 days at very low levels (Fig. 4),
but within the range of sensitivity of two exquis-
itely penicillin-susceptible organisms, group A
Streptococcus and T. pallidum. Moreover, with
monthly intramuscular doses of 1.2 million units,
we protected rheumatic subjects virtually com-
pletely against recurrent rheumatic fever.

Syphilis. Although I hankered to attack
S5T)hilis as well with what seemed to be a single-
injection treatment, I had to be satisfied with re-
porting my results to one of my mentors. Dr.
Harry Eagle at the U.S. Public Health Service
Venereal Diseases Research Laboratory (VDRL).
He promptly confirmed our prolonged blood level
data and the U.S. Public Health Service went
pell-mell after syphilis. The impact on syphilis
was spectacular. Early VDRL data (Table 2)
reveal that primary, secondary, and tertiary
syphilis were cured with an effectiveness equal to
that of any alternative penicillin regimen. One
dose of 2.4 units intramuscularly was all that was
necessary to cure primary or secondary syphilis,
and three weekly doses sufficed for tertiary syph-
ilis, with some rare exceptions. Some of us will
recall the marvel of curing even a huge gumma
with modest doses of penicillin. Those of us who
were once authoritative in the treatment of sj^ph-
ilitic heart disease are now rarely solicited for our
expertise. Although syphilis still lurks all too fre-
quently in the morass of current sexually trans-
mitted infection, and runs wild in patients with
AIDS, the organism remains eminently treatable
with low-dose penicillin regimens.

Rheumatic Fever and Streptococcal Pharyn-
gitis. The studies at Irvington House that I initi-
ated with Bicillin (12), so faithfully continued and
completed by my colleagues Angelo Taranta,
Harrison Wood, and Alvan Feinstein (13), are
summarized in Table 3. Not only has monthly
benzathine penicillin been the most effective reg-
imen for preventing rheumatic recurrences, but
its study has enabled us to learn the natural his-

Fig. 3. Electronmicrophotograph of a group A streptococcus
adhering to receptors on the surface of buccal mucosal cells (E)
(Ref 9).

tory of rheumatic heart disease when subsequent
streptococcal infection is fully controlled in the
rheumatic host.

Moreover, following the seminal studies of
Charles Rammelkamp and his colleagues (14) at
Warren Air Force Base in the late 1940s and
early 1950s, our studies of mass prophylaxis with
Bicillin at Great Lakes Naval Training Center

(15) during my tenure at Northwestern Univer-
sity in Chicago showed how dramatically we
could terminate raging epidemics of rheumatic fe-
ver in military populations. In this typical M19
outbreak (Fig. 5), only three patients developed
acute rheumatic fever after mass prophylaxis was
initiated. Ironically, the three patients were not
given Bicillin because of a vague history of "pen-
icillin allergy." Note, however, that all cases of
streptococcal pharyngitis on this base were not
eradicated from this population; the prevalent
virulent M-type 19 strain disappeared, and with
it, rheumatic fever.

A simplistic concept of rheumatic fever and
group A streptococcal pharyngitis had emerged in
the 1960s. Students learned that rheumatic fever
was, in all of its features, a poststreptococcal in-
fection. But it was already clear to me from my
work and the work of others that not all group A
streptococcal infection could cause rheumatic fe-
ver. The facts were that the epidemiology of group
A infection had dramatically changed. Our stud-
ies at Children's Memorial Hospital, Chicago,
with the late Alan Siegel and with Eloise Johnson

(16) showed that GrAS infection of the throat that
was mild and untreated did not result in rheu-
matic fever (17).

These studies aroused much controversy.
Were these patients mostly "carriers"? If so, why
did those in whom we demonstrated prospectively

116

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

PENICILLIN BLCWO LEVELS

SERUM CONCENTRATION OF PENICILLIN FOLLOWING
SINGLE INTRAMUSCULAR INJECTON OF 600P00
UNITS OF

PENICILLIN G
• PROCAINE PENICILLIN G

PROCAINE PENICILLIN IN OIL WITH 2%ALUM
MONOSTEARATE

BENZATHINE PENICILLIN G

04 RANGE OF PENICILLIN SENSITIVITY
OF GROUP A STREPTOCOCCI

'•■.A...

6 7
DAYS

Fig. 4. Mean serum levels of penicillin in patients who re-
ceived one I.M. injection of 600,000 U of aqueous penicillin G
compared with single injections of three different repository
penicillin salts (Ref 11).

clear ASO rises not develop RF? Our numbers
were considered too few for firm conclusions to be
drawn and ethics committees frowned on further
controlled studies.

Meanwhile, the virulent strains of strepto-
cocci that I called "rheumatogenic" were gone
from U.S. populations. In the Third World's in-
dustrializing countries, however, rheumatic fever
abounded (18). And with the disappearance of
rheumatogenic strains from developed countries,
along went rheumatic fever. Rheumatic fever de-
clined spectacularly in Chicago in the 1970s (Fig.
6) within a period too brief to invoke genetic, eco-
nomic, or any other socioeconomic changes in the
Chicago population as a convincing explanation
(19). Right up into the 1980s, studies of strepto-
coccal sore throat in children showed the same
frequency as always, but without any trace of

TABLE 2

Early Results in the Venereal Diseases Research Laboratory
of the U.S. Public Health Service on Therapy of Syphilis
with Two Forms of Penicillin

Stage

Drug

Dose X 10«

% Relapse

PAM

1.2

12.8

PAM

2.4

6.5

PAM

4.8

2-5.0

Benz

2.4

2-^.0

PAM

4.8

5.0

Benz

2.4

4.0

PAM or Benz

6-9

(2-3 wks)

PAM, penicillin in aluminum monostearate; Benz, benzathine
penicillin.

rheumatic fever (20). Some of my infectious dis-
ease colleagues attributed this situation to good
penicillin therapy, but few patients, if any, re-
ceived benzathine penicillin G and few oral regi-
mens ever commanded compliance with the IC
days of continuous treatment required to preveni
primary attacks of rheumatic fever during epi-
demics caused by rheumatogenic GrAS. Even sec-
ondary rheumatic attacks disappeared, despitt
clear serologic evidence of intercurrent strepto-
coccal infection in the patients with rheumatic
heart disease that Alan Bisno and I followed ir
Memphis (21).

Rheumatogenic strains, I insisted, were gone
What are such strains? As the late Lewis Wan
namaker and others had pointed out, they were
not the GrAS that cause pyoderma. Our studies ir
Memphis with Alan Bisno confirmed that (22). Bj
1970, we clearly demonstrated the sharp separa
tion of the epidemiology of poststreptococcal acute
glomerulonephritis and acute rheumatic fevei
(Fig. 7). Moreover, during the hot months wher
rheumatic fever was gone, GrAS pharyngitis due
to so-called "skin strains" was prevalent. In the
Warren Air Force Base epidemics of rheumatic
fever, several streptococcal M types were associ
ated with the disease, each with the same rheu
matic fever attack rate. From this observatior
grew the monolithic concept that rheumatic fevei
was not M-type dependent.

However, only a limited number of M types
were demonstrated to cause rheumatic fever ir
that and other military populations. Moreover
these strains were highly virulent, encapsulatec
variants which formed mucoid (hyaluronic-acid
encapsulated) colonies. They were rich in M pro
tein (Fig. 8). They were quite different from the
pearly smooth strains isolated in sporadic cases o:
strep throat in civilian populations with no cases
of RF. Such strains may still be M typable anc
may even belong to the common rheumatogenic
M serotypes (23). When they dissociate and lose
virulence, however, they lose rheumatogenicity
Virulence can then be recovered, if at all, only b}
intensive mouse or perhaps human passage.

Most important, rheumatogenic strains have
been shown by Edwin Beachey (24) and confirmee
by Vincent Fischetti at the Rockefeller Univer
sity (25) to have a greatly extended M proteir
molecule of repeating peptide units. This pep-W
extract has been purified and the 14-unit peptide
responsible for typability has been synthesizec
free of cross-reactivity with heart tissue. A mul-
tivalent vaccine for multiple rheumatogenic M
types is potentially produceable, much as has

Vol. 60 No. 2

PENICILLIN— STOLLERMAN

117

TABLE 3

Summary ofFive-Year (1954-1959) Irvington House Study of Three Prophylactic Regimens to Prevent Recurrence of

Rheumatic Fever

BicillinT

Ural renicillint

buliadiazines

Total

Patient Years

560

545

576

1,681

Strep iniections

(Rate per pt. yr.)

Total

6.0

20.6

23.2

16.6

Confirmed*

4.3

18.5

17.7

13.5

Recurrences

Number

Rate

% per pt. yr.

0.4

5.5

2.8

2.9

% per strep infect.

5.9

25.5

11.6

16.5

% per confirmed* strep infect.

8.3

28.7

15.7

20.7

* Antibody rise.

t Benzathine penicillin 1.2 million units IM once monthly.
1: 250,000 units daily.
§ 1.0 g daily.

been accomplished for the multivalent pneumo-
coccal vaccine (Fig. 9).

Will we need a multivalent vaccine in the
United States? The sudden outbreak of rheumatic
fever in the mid-1980s in focal areas of the United
States was startling. The outbreaks in military
personnel at San Diego Naval Base and Fort
Leonard Wood, Missouri (26), and those in civil-
ian communities as far apart as Utah (27) and
Pennsylvania occurred among middle-class
school children who were getting usual medical
care. At the most only a third of the latter had
reported a previous sore throat severe enough to
have made them seek medical treatment. Most

BENZATHINE PENICILLIN PROPHYLACTIC PROGRAM 1958-1959

BFLU

ADENOVIRUS .

': MASS / :

JULY 58 AUG SEPT OCT NOV DEC JAN '59 FEB

Fig. 5. Mass prophylaxis of rheumatic fever (RF) in a popu-
lation of naval military recruits given 1.2 million U of benza-
thine penicillin G during an epidemic of type 19 group A strep-
tococcal pharyngitis (Ref 15).

intriguing has been the return of the famous
rheumatogenic mucoid strains in these communi-
ties, especially M18, a notorious villain in World
War II outbreaks in military populations.

The Fort Leonard Wood epidemic (Fig. 10)
was terminated with benzathine penicillin mass
prophylaxis and the MIS strain seems to have
gone from that base. So has rheumatic fever, but
not endemic streptococcal pharyngitis. In a recent
report, a military health team bravely tried to

CASES
400 -|

1961 1963 1%5 1967 1969 1971 1973 1975 1977
YEARS

Fig. 6. Decline in number of patients with rheumatic fever
or rheumatic heart disease registered annually in Chicago
Board of Health's rheumatic fever registry between 1965 and
1977.

118

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

SEASONAL DISTRIBUTION OF ACUTE RHEUMATIC FEVER (ARF)
AND ACUTE CLOKERULONEPHRITIS (AGN) ADMISSIONS AT
CITY OF MEMPHIS HOSPITALS, SEPT., 1965 - AUG.. 1968

SPRING SUMMER

FALL

Fig. 7. Dissociation of epidemiology of acute rheumatic fever
(ARF) and acute glomerulonephritis (AGN) in southern
United States.

eradicate all cases of strep throat at one military
base by vigorous use of benzathine penicillin
therapy. They have also treated "penicillin hyper-
sensitive" subjects with erythromycin. Rheu-
matic fever, however, disappeared from the base
before these efforts were made. Why? Unfortu-
nately, microbiological studies of the infecting
pharyngeal strains on this base, including M-typ-
ing and colony morphology, were not made. Over-
zealous erythromycin treatment of GrAS sore
throat, as in the 1970s in Japan (19), has caused
the emergence of streptococci multiply-resistant
to erythromycin, chloramphenicol, and tetracy-
cline, but not to penicillin. The speed with which
mutant strains emerge in a population under the
selective pressure of antibiotics is awesome.

SEQUENCE OF M24 - CB7

Fig. 9. Purification of
M24 protein: Amino acid
sequence of a small pep-
tide containing the type-
specific epitope (Ref. 24).

Fig. 8. Mucoid strain of rheumatogenic type M24 group A
streptococcus grown in type-24 antiserum. Electronmicropho-
tograph showing precipitated Ag/Ab complexes around a large
hyaluronate capsule.

Penicillin-Resistant Gram-Positive Organ-
isms and Overzealous Cephalosporin Therapy.

The selective pressure of intensive antibiotic us-
age is evident in the impact of the promiscuous
use of third-generation cephalosporins, which is
now becoming grimly apparent. Simplistic dia-
grams featured in drug-company-sponsored sym-
posia show cephalosporins prepared to lick any
kind of bacterium found in the human host
(whether good or bad for your health). To our hor-
ror, the multiresistant strains of pneumococci,
such as the one that arose out of the pneumonia-
riddled pediatric hospitals of Durban, S. Africa,
are slowly but surely finding their way around
the world (28).

- Fort Leonard Wood, Missouri, 1987 and 1988

Fig. 10. Hospitalization for acute respiratory disease (ARD)
(curve, left ordinates) and present recovery of group A strep-
tococci (GABHS) (upper curve, right ordinates) among Army
trainees at Fort Leonard Wood, Missouri, July 1987 to April
1988. Thirteen cases of acute rheumatic fever were diagnosed
between October 1987 and February 1988. No new cases had
been reported since the second week of March when benza-
thine penicillin G was given once to all nonallergic soldiers
and thereafter to new trainees on arrival at the base.

Vol. 60 No. 2

PENICILLIN— STOLLERMAN

119

As long as we continue to teach our students
to treat pneumonia with shotgun therapy for com-
munity-acquired infections as though they were
treating disease-compromised patients in the in-
tensive care unit, pneumococcus mutants that
can defy penicillin therapy will grow and prosper
in the community. Sensible doses of penicillin
should be used first when circumstances do not
demand an urgent shotgun approach of broad-
spectrum therapy.

Finally, I share with the British poet Hilaire
Belloc these sentiments toward mine enemy who,
unlike me, seems to grow no older, only more de-
vious:

He prayeth best who loveth best.
All things both great and small.
The streptococcus is the test,
I love him least of all.

References

1. Chain E. Penicillin and beyond. (Commentary.) Nature

1991; 353:493^94.

2. King FH, Schneierson SS, Sussman ML, Janowitz HD,

Stollerman GH. Prolonged moderate dose therapy ver-
sus intensive short term therapy with penicillin and
caronamide in subacute bacterial endocarditis. J Mt
Sinai Hosp (now Mt Sinai J Med) 1949; 16(1):35^6.

3. Eagle H, Musselman AD. The rate of bactericidal action of

penicillin in vitro as a function of its concentration and
its paradoxically reduced activity against certain or-
ganisms. J Exp Med 1948; 88:99.

4. Wood WB Jr, Smith MR. An experimental analysis of the

curative action of penicillin in acute bacterial infec-
tions: I. The relationship of bacterial growth rates to
the anti-microbial effect of penicillin. J Exp Med (Bal-
timore) 1956; 103(41:487-498.

5. Schlichter JG, MacLean H. A method for determining the

effective therapeutic lead in the treatment of subacute
bacterial endocarditis with penicillin: a preliminary re-
port. Am Heart J 1947; 34:209.

6. Stollerman GH, Roston EH, Toharsky MA. A guide to the

use of procaine penicillin in hospital practice. NY State
J Med 1948; 48:2501-2505.
7a. Thomasz A. The mechanism of the irreversible antimi-
crobial effect of penicillins: how the beta-lactam anti-
biotics kill and lyse bacteria. Ann Rev Microbiol 1979;
33:113.

7b. Tipper DJ, Strominger JL. Mechanism of action of pen-
icillins: a proposal based on their structural similarity
to acyl-D-alanyl-D-alanine. Proc Nat Acad Sci USA
1965; 54:133.

8. Allan JD, Eliopoulos GM, Moellering RC Jr. The expand-

ing spectrum of beta-lactam antibiotics PBP's. In: Stol-
lerman GH et al., eds. Advances in internal medicine.
Chicago: Year Book Publishers, 1986; 31:119.

9. Beachey EH, Ofek I. Epithelial cell binding of group A

streptococci by lipotheichoic acid on fimbriae denuded
of M protein. J Exp Med 1976; 143:759.
10. Ofek I, Beachy EH. Bacterial adherence. In: Stollerman

GH, et al., eds. Advances in internal medicine. Chicago:
Year Book Publishers, 1980; 25:503.

11. Stollerman GH, Rusoff JH. Prophylaxis against group A

streptococcal infections in rheumatic fever patients.
Use of a new repository penicillin. JAMA 1952; 150:
1572.

12. Stollerman GH, Rusoff JH, Hirshfield I. Prophylaxis

against group A streptococci in rheumatic fever: the use
of single monthly injections of benzathine penicillin G.
N Engl J Med 1955; 252:787-792.

13. Wood HF, Stollerman GH, Feinstein AR, et al. A con-

trolled study of three methods of prophylaxis against
streptococcal infection in a population of rheumatic
children. N Engl J Med 1957; 257:394.

14. Rammelkamp GH, Denny FW, Wannamaker LW. Studies

on the epidemiology of rheumatic fever in the armed
services. In: Thomas L, ed. Rheumatic fever. Minneap-
olis: University of Minnesota Press, 1952:72-89.

15. Frank PF, Stollerman GH, Miller LF. Protection of a mil-

itary population from rheumatic fever. JAMA 1965;
193:775-783.

16. Siegel AC, Johnson EE, Stollerman GH. Controlled stud-

ies of streptococcal pharyngitis in a pediatric popula-
tion: I. Factors related to the attack rate of rheumatic
fever. N Engl J Med 1961; 265:559.

17. Stollerman GH, Siegel AC, Johnson EE. The variable ep-

idemiology of streptococcal disease and the changing
pattern of rheumatic fever. Mod Concepts Cardiovasc
Dis 1965; 34:45-48.

18. Stollerman GH. The relative rheumatogenicity of strains

of group A streptococci. Mod Concepts Cardiovasc Dis
1975; 44:35.

19. Stollerman GH. Global changes in group A streptococcal

diseases and strategies for their prevention. In: Stoller-
man GH, et al., eds. Advances in internal medicine.
Chicago: Year Book Publishers, 1982; 27:373.

20. Holmberg SD, Faich FA. Streptococcal pharyngitis and

acute rheumatic fever in Rhode Island. JAMA 1983;
2.50:2307.

21. Bisno AL, Pearce lA, Stollerman GH. Streptococcal infec-

tions which fail to cause recurrences of rheumatic fever.
J Infect Dis 1977; 136:278-285.

22. Bisno AL, Pearce I A, Wall HP, Moody MD, Stollerman

GH. Contrasting epidemiology of acute rheumatic fever
and acute glomerulonephritis. Nature of the antecedent
streptococcal infection. N Engl J Med 1970; 283:561.

23. Stollerman GH. Rheumatogenic group A streptococci and

the return of rheumatic fever. In: Stollerman GH, et al.,
eds. Advances in internal medicine. Chicago: Year
Book Publishers, 1990; 35:2.

24. Beachey EH, Stollerman GH, Chiang EY, et al. Purifica-

tions and properties of M protein extracted from group
A streptococci with pepsin: covalent structure of the
amino terminal region of type 24 M antigen. J Exp Med
1977; 145:1469.

25. Bessen D, Jones KF, Fischetti VA. Evidence for two dis-

tinct classes of streptococcal M protein and their rela-
tionship to rheumatic fever. J Exp Med 1989; 169:269-
298.

26. Centers for Disease Control. Acute rheumatic fever

among Army trainees — Ft. Leonard Wood, MO, 1987-
1988. MMWR 1988; 37:519.

27. Centers for Disease Control. Acute rheumatic fever: Utah.

MMWR 1987; 36:108.

28. Stollerman GH. The return of rheumatic fever. Hosp

Pract 1988; 23:100.

Editorial

Did the Patient Tell Us?

With this issue the Journal embarks on a new feature, a form of pathological-clinical corre-
lation called here "Autopsy and Medical Care." The classical clinical-pathological conference
is formatted to test the skill, expertise, and perspicacity of the discussant in choosing one or
more differential diagnoses based on history, physical examination, and laboratory findings.
The conference generally concludes with the findings presented by the pathologist and perhaps
some additional remarks by the participants. Cases are usually selected to provide information
of special interest with respect to the gamut of interesting if not unusual differential diagnoses.

The approach in this feature will be somewhat different. Here, the findings by the pathol-
ogist have been given to the discussant. The task of the discussant is to reconstruct the patient's
history and clinical course. With this format, the discussant is forced to focus on the meaning
and significance of each or most of the pathologist's findings and how these might have
influenced the patient's chief complaint, history, physical findings, and results of various
special examinations and laboratory tests. In some ways, this format is analogous to the task
in forensic medicine, where little or no prior information may be available other than the
findings at autopsy. Readers are asked to conjure up their own fiction and test their imagi-
nation and creativity against that of the discussant. At the end of the hypothetical reconstruc-
tion, the patient's true clinical story is documented to permit a comparison of fact and suppo-
sition.

The objective of this exercise is twofold. Attention will have to be paid to all of the findings
at autopsy, as well as how well the true clinical story and course, as recorded by the physicians
caring for the patient, are reflected in these findings. It is hoped that this effort will hone the
skills of taking and recording a history, sharpen insight into the relevance of such a history and
the patient's clinical course, and enhance our profession's use of the findings at autopsy as an
important objective assessment of the quality of care. This being the case, an effort to increase
the autopsy rate must be made. Its present level as practiced in major medical centers is
shameful. — Sherman Kupfer

120

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 19!

Autopsy
and

Medical Care

Clinical Correlates of
Pathologic Findings:

Case 1

Robert E. Siegel, M.D.

Autopsy: Clinical Information

• A 73-year-old black man with a 45 pack-year
smoking history

• History of emphysema and gout

• Status post anterior resection for rectal adeno-
carcinoma in 1987; treated with postoperative
radiotherapy

• Status post transurethral resection of his pros-
tate for adenocarcinoma in 1989; treated with
radiotherapy in 1990 and estrogen in 1991

• Last hospital admission 8/27/91; died 9/6/91

Autopsy:
Pathological Diagnoses

Principal

I. Status post prostatic adenocarcinoma

a. Extensive tumor emboli within pulmonary
vessels

b. Micrometastases, lung parenchyma and
mediastinal lymph nodes

c. Metastatic tumor nodule, liver

II. Status post resection with colostomy for rectal
adenocarcinoma

a. Multiple peritoneal adhesions

From the Medical Intensive Care Unit and the Pulmonary
Section, Bronx Veterans Affairs Medical Center, and the De-
partment of Medicine, Mount Sinai School of Medicine. Ad-
dress all correspondence and reprint requests to Dr. Robert E.
Siegel, Associate Chief, Pulmonary Section, Bronx VA Medi-
cal Center, Bronx, N.Y. 10468.

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 1993

Secondary

• Acute bronchopneumonia, focal — left upper
and lower lobe, right lower lobe

• Pulmonary emphysema, mild

• Arterio-arteriolo-nephrosclerosis, kidneys, mild

• Diverticulosis, colon

Autopsy: Gross Description

External. The body is that of a well-devel-
oped cachectic black male appearing the stated
age of 73 years, measuring 6'1" in length and
weighing approximately 140 lbs. Lymph nodes
are unremarkable. The breasts and nipples are
normal.

Internal. The pleurae are unremarkable and
contain 30 cc. of clear serous fluid. The pericar-
dial sac transverse diameter is 15 cm and con-
tains 100 cc of clear serous fluid. The diaphragms
are at the 7th intercostal spaces anteriorly. The
peritoneum is remarkable for multiple adhesions
and contains a minimal amount of clear yellow-
brown fluid. The liver edge is 2 cm below the cos-
tal margin at the right mid clavicular line.

Cardiovascular System. The heart weighs
490 g (normal 270-360 g). It has a normal config-
uration and the following measurements (normal
values in parentheses):

Tricuspid valve 11.5 cm (10-12.5)

Pulmonary valve 9 cm (7-9)

Aortic valve 9 cm (6-7.5)

Mitral valve 10.5 cm (8-10.5)

Right ventricular thickness 0.5 cm (average 0.5)

Left ventricular thickness 1.5 cm (average 1.5)

121

122

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1995

The myocardium is unremarkable and the aorta
has mild atherosclerotic changes. The coronary
arteries show patchy calcification and atheroscle-
rotic changes with significant narrowing of lu-
mens up to 75%. The superior and inferior vena
cavae and portal veins are unremarkable.

Respiratory. There is no pneumothorax
present. The right lung weighs 400 g (normal
360-570) and the left 360 g (normal 325-480).
The pleural surfaces are unremarkable, with a
moderate amount of anthracotic pigment. Except
for mild emphysema, the parenchyma seems un-
remarkable. The trachea and bronchi appear nor-
mal and the pulmonary vessels are patent and
unremarkable.

Hepatobiliary System. The liver weighs 1320
g (normal 1500-1800). The appearance is normal.
There is a well-circumscribed yellowish-white
firm tumor nodule measuring 1.9 cm in diameter
1 cm deep to capsular surface. The portal veins,
hepatic artery, and portal areas are unremark-
able.

Genitourinary System. The right kidney
weighs 50 g and the left 170 g (normal 125-170
each). The capsules strip with difficulty revealing
an irregular red-brown surface with fetal lobula-
tions and no scars. The prostate is small, very
firm, and yellowish-white in color.

All other systems were unremarkable.

Autopsy:
Microscopic Examination

Lungs. Extensive tumor emboli were found
within pulmonary vessels throughout all lobes of
the lung bilaterally, occasionally obliterating
vascular spaces. Some of these vessels have un-
dergone reactive sclerosis and/or fibrosis due to
complete embolic occlusions. Occasional mi-
crometastases are found in lung parenchyma
within alveolar spaces. Foci of acute broncho-
pneumonia with polymorphonuclear cell infil-
trate within alveolar spaces are seen in sections
of the left upper, left lower, and right lower lobes.
Foci of bronchitis are also seen. Mild emphysema-
tous changes are noted diffusely.

Mediastinal Lymph Nodes. Micrometastases
with occasional giant cells are found in a medias-
tinal node.

Liver. A well-circumscribed metastatic tu-
mor nodule with areas of necrosis is present
within the liver parenchyma. The tumor is a mod-
erately differentiated adenocarcinoma. Immuno-
staining for prostatic specific antigen and prostat-
ic acid phosphatase is strongly positive. The

remaining liver parenchymal architecture is in-
tact with diffuse congestion and accumulation ol
bile pigment within occasional hepatocytes and
canaliculi.

Heart. Multiple sections show moderatelj
hypertrophied cardiac myocytes.

Kidneys. Scattered sclerosed glomeruli an
seen in the kidneys bilaterally, more severely ir
the right. There is thyroidization of multiple tu
bules with accumulations of hyaline within tubu
lar lumens. Diffusely thickened arterial walls
with atherosclerotic changes are observed. Then
is a mild interstitial inflammatory cell infiltrate
with mild congestion.

Prostate. Sections through the prostate re
veal extensive fibrosis with no residual tumo]
present.

Large Intestine. Sections of colon show diver
ticulosis and no residual tumor.

All other organs were unremarkable.

Hypothetical Reconstruction of
Clinical Case from
Autopsy Findings

The following case history is reconstructed or
the basis of the pathological findings delineatet
above. I have included expected laboratory am
hemodynamic parameters and a probable clinica
management of the patient. I discuss the patho
physiologic basis for the expected clinical find
ings. Finally, I discuss my reconstruction of th(
diagnostic studies, pathogenesis, and treatmen
of this patient's disease process with reference t(
significant clinical studies.

I would guess that a few years ago, the pa
tient began to experience mild shortness of breatl
going up stairs as a result of his emphysema. Tw(
to three weeks prior to this admission, I think h(
started to experience increased difficulty carrying
packages due to dyspnea and the sensation of aii
hunger during basic activities of daily living, in
eluding bathing and shaving. His symptomi
would have progressed until he felt breathless a
rest, prompting him to seek the help of his phy
sician and admission to the hospital. Episodii
dyspnea progressing to dyspnea at rest is a com
mon complaint of patients with tumor emboli oi
pulmonary hj^ertension.

Hypothetical Examination and Study Find
ings. On physical examination, he appearec
cachectic. He would be in moderate respirator]
distress, with a respiratory rate in the range o
25-30/min. His blood pressure might have beer
elevated to 160/90 mm Hg with a regular puls(

Vol. 60 No. 2

AUTOPSY AND MEDICAL CARE— SIEGEL

123

rate of 120/min. The jugular veins would be dis-
tended with the patient sitting upright. The lungs
would be clear on auscultation. Cardiac examina-
tion would reveal a loud P2 and an S3 which in-
creased with inspiration. Abdominal examination
revealed a 15-cm pulsatile liver without a palpa-
ble spleen. There would be no clubbing of the fin-
gers or cyanosis but there would be 2 -I- edema of
the lower extremities. Roman's sign was nega-
tive. A chest film would reveal borderline cardiac
enlargement and clear lung fields. I would have a
high degree of suspicion of pulmonary emboli
given the clear lung fields, rapid onset of dyspnea,
history of malignancy, and estrogen use.

The patient should have a high minute ven-
tilation due to extensive emboli. Although the
physical blockage of pulmonary arterioles and
capillaries by tumor causes alveoli to be venti-
lated and not perfused, release of mediators and
bronchospasm lead to hypoxemia. In addition, the
plugging of capillaries with microscopic tumor
emboli can cause a diffusion abnormality which
can add to hypoxia. This patient should have all
of the expected physical signs of pulmonary hy-
pertension and right ventricular failure due to a
significant reduction in the size of the functional
pulmonary capillary bed. This is seen only with
extensive pulmonary emboli, since the pulmo-
nary vascular bed is a low-pressure, highly com-
pliant circuit. Clinical appearance includes clear
lungs, since the disease is confined to the vascu-
lature and spreads to the lymphatics. If there is
further spread to the bronchi, it is possible that
one might hear a localized wheeze or feel a pal-
pable rhonchus over a nearly obstructed bron-
chus.

In the emergency room, the results of an ar-
terial blood gas test would have been pH7.48,
PCO2 26mm Hg, PO2 48mm Hg and oxygen satu-
ration of 84%. These findings would indicate re-
spiratory alkalosis as a result of tumor emboli
and metabolic acidosis secondary to lactic acid
and hypoxia.

The patient would be anticoagulated with
heparin for treatment of pulmonary emboli.

Other significant laboratory findings would
include: hemoglobin of 10.4 g/dL, white blood cell
count of 12,000/uL, alkaline phosphatase of 235
U/dL, SCOT of 145 U/L, SGPT of 200 U/L, lactic
dehydrogenase of 250 U/L, and total bilirubin of
1.5 mg/dL. ECG was significant for a sinus tachy-
cardia at 120 and a right bundle branch block
(RBBB). Lung perfusion scan revealed multiple
subsegmental peripheral perfusion defects. A pul-
monary arteriogram was normal.

Heparin treatment was continued due to

high clinical suspicion of pulmonary emboli, and
broad-spectrum antibiotic therapy was begun to
cover possible sepsis syndrome.

Elevated liver function tests reflect tumor
emboli to liver and passive congestion from right
ventricular failure. Tachycardia is secondary to
increased oxygen consumption from increased
work of breathing. Tachycardia is also the conse-
quence of falling left ventricular stroke volume,
resulting from decreased left ventricular preload
due to right ventricular failure. High right ven-
tricular pressure from pulmonary hypertension
can cause decreased diastolic compliance of the
left ventricle due to septal shift and reduced left
ventricular stroke volume. The right bundle
branch block is due to high right ventricular pres-
sure effect on the superficial subendocardial right
bundle of His. Pulmonary angiogram usually will
not demonstrate emboli, since tumor emboli in-
volve small arteries and arterioles, which are
poorly evaluated by this test. A normal pulmo-
nary angiogram would rule out significant throm-
boemboli, however, unless it is a technically poor
study.

After thromboembolic disease was ruled out,
the patient's clinical picture of tachycardia, leu-
kocytosis, hypoxia, and abnormal results of liver
function tests suggested sepsis syndrome. Al-
though the chest film was normal, it was sus-
pected that the patient's hypoxia was a result of
early adult respiratory distress syndrome ( ARDS)
as a component of multiorgan failure. A diffuse
alveolar filling process was expected to follow in
subsequent films.

Hypothetical Clinical Course. The patient
was transferred to the medical intensive care
unit, where he deteriorated over the next 24—72
hours. Cardiac enzyme levels were normal and
blood cultures were negative. Echocardiogram
was performed to rule out tamponade as the cause
of RV failure. Echocardiogram revealed a small
pericardial effusion, tricuspid regurgitation, di-
lated right atrium and RV and good LV function.

As the patient developed worsening RV fail-
ure and h5q)oxia, a pulmonary artery (PA) cathe-
ter was inserted, so that hemodynamic parame-
ters could be monitored and appropriate
interventions made. Hemodynamic study re-
vealed the following values: central venous pres-
sure of 25 mm Hg (normal 2-8); pulmonary artery
pressure of 62/36 mm Hg (normal systolic pres-
sure 15-30 mm Hg; normal diastolic pressure
4—14 mm Hg); mean pulmonary pressure of 45
mm Hg (normal 8-18) and wedge pressure of 6
mm Hg (normal 2-12). The cardiac index was 1.8
1/min/m^ (normal 2.4—3.8), and the systemic vas-

124

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

cular resistance 2014 dyn • sec/cm^ (normal 800-
1500). Pulmonary vascular resistance was 962
dyn • sec/cm^ (normal 100-300). High right-sided
pressures and decreased cardiac output reflect
right ventricular failure. Increased systemic vas-
cular resistance is secondary to a hyperadrener-
gic state from falling cardiac index.

An increased serum lactate of 3.8 mEq/L
(normal <2.0) reflected an inadequate oxygen up-
take by the tissues. This patient had increased
oxygen supply needs due to increased work of
breathing. He also suffered from low oxygen de-
livery to tissues due to a combination of low car-
diac index and low arterial oxygen content of
blood (oxygen delivery = cardiac index x arterial
oxygen content).

His respiratory status gradually deteriorated
over the next few days. Despite a 100% oxygen
mask, his respiratory rate increased to 40 per
minute with hypoxia and he was intubated for
impending respiratory failure. He was treated
with volume loading and amrinone, with little
improvement in pulmonary artery pressures or
cardiac index. In light of a decrease in cardiac
output from right ventricular failure, amrinone
was the drug of choice for this patient. It is a
myocardial inotrope and a vasodilator due to in-
hibition of phosphodiesterase activity. This agent
is more effective at vasodilation of the pulmonary
circuit then dobutamine, which vasodilates via p2
stimulation. When there was a failure of stroke
volume to improve with amrinone, dobutamine
was added, with some improvement in stroke vol-
ume. If amrinone is ineffective at increasing right
ventricular function, dobutamine can be added,
since it acts via a different mechanism, stimulat-
ing p receptors to increase ventricular function.

Bronchoalveolar lavage was performed to
look for infectious agents and tumor. Neither
Pneumocystis carinii nor malignant cells were
found. Tumor emboli were considered likely in
this patient, since hypoxemia without diffuse in-
filtrates would make the diagnosis of ARDS un-
likely. Blood withdrawn from the distal tip of the
pulmonary artery catheter in the wedged position
was sent for cytologic examination and was re-
ported positive for malignant cells. Cytologic
examination of specimens obtained from either
pulmonary artery capillary blood or from bron-
choalveolar lavage can be useful in demonstrat-
ing invading cells. The patient developed in-
creasing hypoxia with hypotension from right
ventricular failure.

In light of the terminal state of the patient
and diffuse metastatic disease, the family decided
not to continue efforts to resuscitate this patient.

The patient was sedated and was incapable of as-
sisting the family with this decision. Ultimately
he became bradycardic and developed electrome-
chanical dissociation.

Discussion of
Hypothetical Reconstruction

This patient died from microscopic pulmo-
nary tumor emboli leading to pulmonary vascular
hypertension, hypoxia, and right ventricular fail-
ure. The initial impression is usually that the pa-
tient has thrombotic pulmonary emboli, since
there is a large overlap in the initial signs and
symptoms of thrombotic and neoplastic pulmo-
nary emboli. Patients with neoplastic pulmonary
emboli have subacute dyspnea, cough, chest pain,
and hemoptysis, in decreasing frequency (1). In a
study by Kane et al., 8 of 16 patients with tumor
emboli of pulmonary arteries and arterioles had
severe, unexplained dyspnea (2). Initial signs in-
clude tachypnea, tachycardia, JVD reflecting
high right-sided filling pressures, clear lungs,
and a loud P2. Elevated liver function tests, re-
flecting metastatic disease, may help to differen-
tiate this entity from thrombotic pulmonary em-
boli (2). A lung perfusion scan may reveal
multiple peripheral defects or a "contour map-
ping" pattern, outlining bronchopulmonary seg-
ments. This pattern reflects small peripheral tu-
mor emboli in patients with tumor microemboli
and lymphangitic spread (3). This is different
from the segmental and subsegmental defects
seen in thromboemboli, which involve larger ves-
sels. Pulmonary angiography is usually normal
but may show slow filling, tortuosity, or pruning
of small peripheral vessels (1).

Kane et al., in an autopsy review of 1085
cases of solid malignant neoplasms over a 15-year
period, found an incidence of isolated microscopic
tumor emboli, not associated with lymphangitic
carcinomatosis, in 2.4%. The most common pri-
mary sites of tumor were prostate and breast,
with liver involvement in 11 of 16 cases and liver
the primary source of tumor in 3 of those 11 (2).

Metastatic tumor involvement of the lungs
can have one of three distinct patterns or a com-
bination of the three:

Microscopic Tumor Emboli. Tumor cells
may be spread hematogenously to peripheral
small arteries and arterioles, causing thrombus
formation and sometimes fibrosis. This is most
common with adenocarcinomas, especially of the
breast or stomach, and is unusually common in
hepatoma. It may occur in up to 10% of patients
with primary lung carcinoma, and occurs more

Vol. 60 No. 2

AUTOPSY AND MEDICAL CARE— SIEGEL

125

commonly in cancers with liver involvement (4).
Patients initially have dyspnea, hypoxia, and
clear chest films. This entity is difficult to differ-
entiate from thrombotic pulmonary emboli. Pa-
tients may develop pulmonary hypertension and
cor pulmonale or die from rapidly progressive
right heart failure.

Lymphangitic Tumor Pattern. After hema-
togenous spread of tumor aggregates, intravascu-
lar tumor can spread through lymphatic channels
into the interstitium, causing a lymphangitic pat-
tern on chest film. Tumor may invade pulmonary
parenchyma, causing nodules on chest film, or
may invade bronchi to cause obstruction and col-
lapse (5). Tumor may also spread medially via
lymphatic channels to hilar and mediastinal
lymph nodes. Neoplastic infiltration can cause
peribronchial thickening and accentuation of in-
terlobular septa. Patients usually have dyspnea.
Chest films reveal increased interstitial mark-
ings, with or without nodules.

Pulmonary Artery Tumor Emboli and In-
farction. Less commonly, tumor may lodge in seg-
mental and larger arteries causing infarction, he-
moptysis, and chest pain or sudden death (1, 6, 7).

Diagnosis of lymphangitic carcinomatosis
can be made by bronchoalveolar lavage (8), trans-
broncheal biopsy of the lung (9), open lung biopsy
or cytologic examination of capillary blood aspi-
rated from a PA catheter wedged in a distal pul-
monary artery (10, 11). It is unclear what the di-
agnostic yield of these procedures is in patients
with clear chest films and microvascular emboli.
In the cytologic study of blood aspirated from a
wedged PA catheter by Masson (11), one of eight
patients described had a clear chest film, and the
diagnosis of metastatic prostate cancer was con-
firmed by both cytologic examination of pulmo-
nary capillary blood and transbroncheal biopsy.

Autopsy. Autopsy of this patient revealed
that the heart was enlarged, with a weight of 490
g (normal 270-360). The myocytes were moder-
ately hypertrophied, yet right and left ventricular
thicknesses were normal. The fact that the right
ventricle was not appreciably thickened, despite
widespread metastatic disease of the pulmonary
vasculature, would imply that there was inade-
quate time for full right ventricular hypertrophy
to accommodate increasing pulmonary artery
pressures, thereby leading to acute right ventric-
ular failure. Acute cor pulmonale is characterized
by dilatation of the right ventricle. In chronic cor
pulmonale there is also hypertrophy of the right
ventricle in response to increased afterload. The
most frequent cause of acute cor pulmonale is pul-
monary emboli. Whereas the lung has significant

thrombolytic potential and fibrinolytic activity
can usually lyse showers of small and medium
size emboli, tumor emboli occlude the vessels
more permanently. The result is a dramatic in-
crease in pulmonary artery pressures in response
to reduction in vascular bed, humorally mediated
vasospasm and bronchospasm with ventilation-
perfusion abnormalities leading to hypoxic vaso-
constriction. If pulmonary pressures rise rapidly,
cardiac output begins to fall as mean pulmonary
artery pressure approaches 30 mm Hg. Evidence
indicates that the acutely loaded right ventricle
fails as mean pulmonary artery pressure reaches
the 40 mm Hg level (12).

Postmortem examination revealed an ab-
sence of residual tumor in the prostate, despite
the fact that the patient died from metastatic
prostate adenocarcinoma. This is not unexpected,
because external radiation is effective in treating
localized tumor (13). The patient most likely had
stage C prostatic cancer (tumor with extracapsu-
lar extension) that had metastasized prior to
treatment with radiation. Metastatic spread of
cancer of the prostate is both lymphatic and he-
matogenous. Metastases occur most frequently to
lymph nodes, followed by bones, lungs, bladder,
and liver (14). However, lung and liver are un-
common as the sole metastatic sites.

References

1. Schriner RW, Jay HR, Edwards WD. Microscopic pulmo-

nary tumor embolism causing subacute cor pulmonale:
a difficult antemortem diagnosis. Mayo Clin Proc 1991;
66:143-148.

2. Kane RD, Hawkins HK, Miller JA, Peter SN. Microscopic

pulmonary tumor emboli associated with dyspnea. Can-
cer 1975; 36:1473-1482.

3. Sostman HD, Brown M, Toole A, Bobrow S, Gottschalk A.

Perfusion scan in pulmonary vascular/lymphangitic
carcinomatosis: the segmental contour pattern. AJR
1981; 137:1072-1074.

4. Fraser RG, Pare JAP, Pare PD, Fraser RS, Genereux GP.

Diagnosis of diseases of the chest. Philadelphia: WB
Saunders 1989, 1641.

5. Janower ML, Blennerhassett JB. Lymphangitic spread of

metastatic cancer to the lung: a radiologic-pathologic
classification. Radiology 1971; 101:267.

6. Winterbauer RH, Elfenbein IB, Ball WC Jr. Incidence and

clinical significance of tumor embolism to the lungs.
Am J Med 1968; 45:271.

7. Gonzalez- Vitale JC, Garcia-Bunnel R. Pulmonary tumor

emboli and cor pulmonale in primary carcinoma of the
lung. Cancer 1976; 38:2105.

8. Levy H, Horak DA, Lewis MI. The value of bronchial

washings and bronchoalveolar lavage in the diagnosis
of lymphangitic carcinomatosis. Chest 1988; 94:1028-
1030.

9. Aranda C, Sidhu G, Sasso LA, Adams FV. Transbronchial

126

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

lung biopsy in the diagnosis of lymphangitic carcino-
matosis. Cancer 1978; 42:1995-1998.

10. Masson RG, Ruggieri J. Pulmonary microvascular cytol-

ogy: a new diagnostic application of the pulmonary ar-
tery catheter. Chest 1985; 88:908-914.

11. Masson RG, Krikorian J, Lukl P, et al. Pulmonary micro-

vascular cytology in the diagnosis of lymphangitic car-
cinomatosis. N Engl J Med 1989; 321:71-76.

12. Miller GAH, Sutton GC. Acute massive pulmonary embo-

lism clinical and haemodynamic findings in 23 patients
studied by cardiac catheterization and pulmonary an-
giography. Br Heart J 1970; 32:518-523.

13. Hanks GE. External beam radiation treatment for pros-

tate cancer: still the gold standard. Oncology 1992; 6;3:
79-89.

14. Saitoh H, Hida M, Shimbo T, Nakamura K, Yamagata J,

Satoh T. Metastatic patterns of prostatic cancer. Cancer
1984; 54:3078-3084.

Actual Clinical Summary

The patient was a 73-year-old black man who
was admitted to the hospital on August 27, 1991.
One month before admission he began to notice
dyspnea on exertion which gradually became
worse until he was short of breath at rest. He
developed a cough productive of small amounts of
white sputum. There was no hemoptysis, chills, or
fever. There was marked weight loss associated
with anorexia.

In 1987, he had a low anterior resection with
a permanent colostomy for adenocarcinoma of the
rectum. He received pre- and post-operative ra-
diotherapy, a total dose of 2000 rads. Two years
later he had a transurethral prostatectomy, at
which time cancer of the prostate was found. His
treatment consisted of local irradiation and estro-
gen medication. In May 1991, a bone scan was
done because of severe back pain. Multiple me-
tastases were demonstrated in lumbar and tho-
racic vertebrae. Additional local radiotherapy to
these areas resulted in amelioration of symptoms.

He had been a cook. He smoked one pack of
cigarettes per day for 45 years, and drank only
occasionally.

On admission there was evidence of cachexia
and bitemporal wasting. He was tachypneic with
a respiratory rate of 26/min. The blood pressure
was 90/60 mm Hg, and the pulse rate was 120/
min. His temperature was 98.8°F. The lungs were
clear on auscultation.

On chest x-ray the lungs were hyperinflated
with a paucity of markings. The diaphragms were

flat. EKG revealed a normal axis with inverted T
waves in leads II, III, and aVf. There was 0.5 mm
depression of the S-T segments in leads V4-6.

Laboratory examination revealed a hemoglo-
bin of 10.6 gm/dL and a hematocrit of 24%. The
white blood cell count was 5,500/mm^. The serum
sodium was 136 mmol/L, potassium 4.0 mmol/L,
chloride 101 mmol/L, and CO2 content 16.4 mmol/
L. The serum creatinine was 1.2 mg/dL, uric acid
9.7 mg/dL, alkaline phosphatase 653 U/L, 7-glu-
tamyl transpeptidase (GOT) 165 U/L, aspartate
aminotransferase (AST, SCOT) 79 U/L, alanine
aminotransferase (ALT, SGPT) 93 U/L, and total
bilirubin 1.1 mg/dL.

The arterial blood gases with the patient
breathing room air were pH 7.50; PO2, 42 mm Hg;
PCO2, 19 mm Hg; oxygen saturation, 84%; with
the patient breathing 100% O2, the arterial blood
gases were pH 7.48; PO2, 141 mm Hg; PCO2, 22
mm Hg; oxygen saturation, 99%.

A VQ scan revealed heterogeneity of both
lungs without discrete defects. The scan was in-
terpreted as indicating low probability for pulmo-
nary emboli. Repeat chest x-rays were un-
changed, as were repeated blood gases. He
received several transfusions and antibiotics. He
was not heparinized. Two days before he died, dif-
fuse rales and rhonchi were heard bilaterally.
During the course of his brief hospitalization, he
became progressively hypoxic despite supplemen-
tal oxygen and died on 9/7/91.

Narrative and Medicine

Oliver Sacks, M.D.

It's a special pleasure being here with you, es-
pecially the students among you, today. I never
Uked residents as much, but I loved students, who
seemed much more unspoilt and open and less
pressured and, in a creative way, less focused. In
my teaching, the neurological rotation was eight
weeks long. We had seven or eight students, and
met them each Friday afternoon. In eight after-
noons you can learn to know people pretty well. I
loved that teaching, and I loved the students; I
think they liked me. Sadly I've had much less to
do with students since about 1974 because of
changes in the medical school. Certainly one of
the things I have been happy to preserve for 20
years is the Alpha Omega Alpha key which you
gave me in 1970. It was a delight to meet you all
informally this afternoon. I wish that meeting
could have gone on for hours and hours, because I
am actually much happier with dialogues than I
am with lecturing, which is a monologue. I like
dialogues. And I feel much more comfortable with
a pen.

Being here today, then, has partly thrown me
back on memories of being a medical student my-
self in London in the late 1950s and how it was
then, at least how it was for me, how it was for
some of us. I think we were probably somewhat
less pressured than you: we didn't have so many
exams, we didn't have grades, there wasn't so
much competition. A lot of us were drunk most of
the while. The medical student of that time has, I
think, a somewhat different stereotype from the
medical student now. For myself, I hated classes,
I hated formal teaching, I didn't absorb any of it,
but I felt very privileged at being free to go to

Adapted from the author's Alpha Omega Alpha lecture at the
Mount Sinai Medical Center on April 30, 1991. From the De-
partment of Neurology, Albert Einstein College of Medicine,
New York. Address reprint requests to the author, Clinical
Professor of Neurology, Albert Einstein College of Medicine at
299 West 12th Street, Apt. 14C, New York, NY 10014.

patients and spend time with them. I've never
been very good at learning from texts or lectures.
My only texts have been individuals, have been
patients.

In particular, I remember one patient who
was an old tea planter from Ceylon in a terminal
uremic delirium, who was babbling all day. His
condition had been diagnosed as uremic enceph-
alopathy. His utterances were delirious and
therefore "nonsense" and "of no interest," and I
was dissuaded from spending time with him. But
I did spend time with him, I listened by the hour,
by the day, and it was an amazing experience,
like being privy to a dream. Gradually some of
the landscapes of his thinking and his recollection
became familiar to me, his symbolism became fa-
miliar to me. And then I started throwing in an
occasional comment myself, and he responded to
that, and this was like talking to a man in a
dream and being both part of his dream and yet
outside the dream. I think it was pleasing to him,
because under the clutter of the delirium I think
he felt very alone and very isolated, and knew
that the end was close. So there was this strange
relationship with a patient who was in a delir-
ium, and for some reason it stays in my mind
more than any other experience in medical
school, and I feel it may have been a sort of pro-
totype for what has happened since.

As medical students, we were, however, en-
couraged and made to tell stories and to give case
presentations. And I loved also listening to the
cases presented by my seniors. I hated lectures, I
loved stories; I still hate lectures, I still love sto-
ries— which is why my so-called lecture may de-
generate into a story. Medicine starts with sto-
ries. The patient has a story. He is a story. He
tells you his story. It may not make much sense to
him, but you help him to fill in the story, you
make connections for him. You bring your own
background, you bring your own cases, and grad-
ually between you the story develops. As you

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 1993

127

128

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

heard earlier, I grew up in an atmosphere of med-
ical stories. My mother, who was a surgeon, loved
telling stories, which would sometimes start with
a surgery or a tumor, or the pathology, but then
spread out to the whole of life and would go on
and on. Her style didn't differ whether she spoke
to her colleagues, her patients or her students,
the milkman or the gardener, or told stories over
the dinner table. I think I partly derived from this
my antagonism to the separation of the academic
and the popular, and my feeling that a story well
told should be accessible and transparent but
should also have all the information one could
want.

I got a little concerned listening to some of
you earlier today and getting a feeling of a time of
life where there might be no time, and much pres-
sure and anxiety and urgency. But as I continued
listening to you, it was also obvious that there
were all sorts of deep and positive attachments.
You would get interested, you would be delighted
by things you heard. You would be drawn into
caring for someone and into caring for phenom-
ena. There is an obvious warmth and attachment
in being a student of medicine, which perhaps one
has more strongly than in any other studenthood.
I was saying to some of you informally a bit ear-
lier, and I want to say it again now, I think that
graduation and residency may represent a sort of
awful bottleneck and squeezing and constriction
of life of a sort which you will never have to en-
dure again. So I think you can have hopes that
things will expand wonderfully later. Certainly
this is what I found myself.

It did seem to me that as medical students in
the lazy, drunken '50s, we had quite a lot of time
before we failed our exams and got talked to very
seriously. I think we often went to plays, and
there were poetry readings, and there were con-
certs. My parents met as medical students at an
Ibsen society for medical students. I think you
need to make time for other things; I think this
is terribly important.

Luria: Scientific Medicine and Narrative. I
took my medical degree at the end of 1958. The
following year, the Russian neuropsychologist
A. R. Luria came to London and he gave some
talks about the development of language, espe-
cially as he had observed it in some identical
twins thirty years before. It was a wonderful talk
in its intellectual depth and brilliance. (Also, this
was the first time I had heard of Chomsky's work
being brought into clinical discourse.) But it was
also beautiful because his affection for these
twins, whom he had seen thirty years before,
shone through his words — one had the feeling of a

man actuated by love. He was a grand anal5rtic
storyteller. When he talked in London, he told
stories, but the stories were full of what he used to
like to call "neuroanalysis," which is the neurol-
ogist's equivalent of psychoanalysis. I was very
pleased that a man like Luria was alive and that
I had a chance of listening to him, because I think
already in the 1950s one felt the power of tech-
nology beginning to close in. There may also have
been a certain hint that clinical medicine and
clinical description were perhaps old hat, were
obsolete, were being replaced by a more scientific
medicine. But Luria himself embodied both scien-
tific medicine and narrative, and these were
brought together in his intense sense of history, of
the unique development and becoming of each in-
dividual.

This was very different from the sense of con-
striction which I got from medical school, and I
think it was partly to escape that sense of bottle-
neck that I bolted from England in 1960 and
hitchhiked around Canada and the States and
then went to California, where I thought every-
thing was very loose — this seemed to me a good
place perhaps to try and disorganize oneself in the
hope that one might then re-organize oneself in a
way that was more personal and more autono-
mous. As residents at UCLA, we had a journal
club which met every week. There was a very big
accent on the latest. I didn't like that; I wish there
had been a history club. But I didn't very much
like medical school or residency, as I've said, and
it was a wonderful relief and surprise when I got
out, and found the world was still there, and that
there was a lovely sort of medicine which could
stimulate and organize and allow one to develop
oneself in a very personal way.

Migraine: The Whole Patient, the Complete
Narrative. The first patients I saw when I got out
of residency were patients with migraine. Mi-
graine sounds dull: headache. But it's not just
headache; it's a huge landscape. It's a wonderful
subject, and it was the subject which first excited
me neurologically, biologically. I had initially re-
garded my patients with migraine in purely neu-
rological terms, but then I was called out of this
attitude by patients telling stories and admitting
me into their lives.

There were two patients in particular who
changed my mind somewhat about things. One of
them was a young man who got migraines every
Sunday. These were severe, classical migraines,
and it was a big event. The family would crowd
around, and icepacks would be put on him, and
there was a lot of sympathy. But strangely he had
never really been given a specific medication for

Vol. 60 No. 2

NARRATIVE AND MEDICINE— SACKS

129

his attacks. So I stepped in and I said, "Let's give
you some ergotamine," which is highly specific, as
you know, or can be. The next Sunday I got a very
excited phone call, saying, "It's wonderful. As
soon as I had the aura, as soon as I saw the zigzag,
I took the ergotamine, the headache didn't de-
velop, and I feel great. You've changed my life."
So that was nice. The next Sunday I didn't get a
call, so I called him later. And in a rather dispir-
ited voice, rather tired voice, he said he had had
the aura, had taken the tablet, the migraine
hadn't developed — but he said he had been sort of
bored all day, he hadn't known what to do. His
Sundays had always been richly filled with mi-
graine; now he didn't know what to do, no one
visited him. And he said he had a premonition, a
foreboding, that something would happen.

The next Sunday I got an emergency phone
call, not from him, but from his sister, saying he
was in the throes of a status asthmaticus. He was
very ill: he was on oxygen, and he was being
given adrenaline, but later that day I talked to
him. I then got a bit of history which I hadn't got
the first time. He said he had had asthma attacks
until he was seven, and then they stopped, and he
had migraine attacks instead. And he added, "I
want them back. I want my migraine back." He
said, "I know where I am with the migraine." And
he said something interesting. He said, "Do you
think I need to be ill on Sundays?" This made me
think in a way in which I had never thought be-
fore. I had simply thought in terms of the patho-
physiology and the pharmacology of migraines, as
a neurologist does think, but he had asked me a
question in terms of needs, in terms of an entire
pattern of a life. To abbreviate the story, I said,
"Well, that's an odd suggestion, let's talk about
it." And we talked about health and illness and
the disposition of a life and whether he might
need to be ill on Sundays and whether there was
some alternative, some nonpathological alterna-
tive, to being ill on Sundays. And as we talked
over a few weeks, his symptoms started to get
less. I suppose this was some of my own discovery
of the "talking cure" and of how essential it is to
look at the whole patient and get a complete nar-
rative and not just a particular symptom.

Case History: Pathography and Biography.
When I was a resident I wrote several articles.
They were articles with orthodox formats, and I
had no difficulty publishing them. They didn't say
much, they didn't provoke much. But when I
started really to deal with patients, I had an ir-
resistible need to tell their stories, and to tell at
least what Hippocrates would call a pathography,
a picture of disease, a story of disease, and some-

times to go on from there to a biography. As soon
as I did that, I found that I could no longer be
published in the medical literature. I think things
have changed somewhat, but certainly in the
1960s it was very difficult to publish a full, indi-
vidual case history. The notion was that science
and clinical science should consist of lots of cases,
and of an average, and the uniqueness of the in-
dividual rather disappeared.

As I had been very excited by listening to
Luria in person when he came to London in 1959,
so I was tremendously excited when in 1968 I
read a book of his called The Mind of a Mnemon-
ist. I started the book and read it for 30 pages
thinking it was another one of those marvelous
Russian novels, and then I realized it was a case
history, a minutely detailed, densely detailed,
systematic case history which somehow had the
feel and the style and the sensibility of a novel. So
Luria had done it again. He had somehow com-
bined science with narrative into a form so right
that one felt any other form would have been
wrong. I was very excited and encouraged when I
read this, and it gave me a feeling that I should
write narratives, and publication be damned. But
it's not easy to say publication be damned if you
write things and they get rejected by medical
journals. In a sense, I was forced by the situation
to try to publish outside of medicine. So really my
first expression of my experience with the Awak-
enings patients was not in Neurology or the Ar-
chives of Neurology but in a general journal in
England called The Listener.

In 1972 Luria published another book, called
The Man with a Shattered World, and I wrote a
review of it which was published on the same day
as Awakenings was. About two weeks afterward I
got a letter, in fact I got two letters, from Moscow.
The envelopes were addressed in a beautiful Vic-
torian handwriting. Getting a letter from Luria
excited me as much as getting a letter from Freud
would have. In one of his letters Luria talked at
length about the beautiful clinical narratives, the
accurate clinical descriptions, of the last century,
especially in neurology and psychiatry, and how
these narratives collapsed or disappeared, or
seemed to, with the rise of technology and statis-
tics and physicalistic models of the organism, of
the nervous system. He said, "The art of clinical
description is almost lost now. It must be re-
vived." I was haunted by those words. I still am
haunted by them.

Artificial Language and Natural Language
in Medicine. Let me jump back 300 years. When
the Royal Society was founded in the 1660s, some
guidelines were laid down for scientific writing. It

130

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

was felt that elaboration, ornament, idiosyncrasy
should be excised, that one should have writing
that was transparent and without personality,
and also writing of a universal sort, so that if it
was translated into Latin or French or Grerman
the question of translation wouldn't matter.

The prescription for scientific writing laid
down in the 1660s was not really a prescription
for prose. I think it was almost a prescription for
an artificial language. I think it is the use of ar-
tificial language which has become so repellent in
medical writing today and sometimes makes it so
opaque and so difficult of access. I don't know to
what extent one needs a special language for
physics and chemistry, but I don't think one needs
a special language for medicine and for describing
human predicaments which all of us know about,
or all of us can imagine. I think that natural lan-
guage is good enough for describing any human
condition, any human condition whatever. And as
physicians what we are concerned with is describ-
ing human conditions. I think that a good narra-
tive, a case history, is not just idle deference to
the past, I think it's not just an archaism, and it's
not just being sentimental or dramatic.

Edelman and the Roots of Consciousness.
This brings me as an ending point to what I think
is one of the most exciting developments in neu-
rology and psychology and biology and neuro-
science at the moment. I'm not referring to all the
wonderful new techniques of brain imaging and
multi-unit recording and things like that, but
rather to the theory of perception and perceptual
categorization and learning and behavior and
language and memory and consciousness which is
being developed at Rockefeller University by Ger-
ald Edelman.

Edelman got the Nobel Prize for his work in
immunology and mechanisms in immune recog-
nition, and he is now turning his enormous
searchlight of a mind on what it means to be hu-
man and what it means to be conscious. What is
emerging is something not at all like the compu-
tational model of the mind or the notion of
thoughts as algorithms, but a view of the nervous
system as always creative, as creating new cate-
gories of perception, of language, of conscious-
ness— the notion of the nervous system as adven-
turous, as facing radical contingencies all the
while and as coming up with new solutions.

The theory of development and of conscious-
ness which comes out of Edelman's work is rather
like a novel. It's full of richness and density and
unexpectedness. When I started to read Edelman,
I breathed an enormous sigh of relief, because it
seems to me in a sense to vindicate and redeem

the narrative and the case history by showing
that the whole is supremely important, that noth-
ing must be left out, and that every patient is
unique and no two stories are the same.

I think in the 1990s we are going to see not
only incredible advances in neurology and neuro-
science, but a renaissance of clinical narrative.
This depends, in part, on young doctors — on their
reviving, in new terms, the ancient art of case
history and clinical narrative. Good clinical nar-
ratives are as valuable as the "hardest" neuro-
science, and indeed its natural complement. So I
hope no graduating medical student will despise
the clinical mode, but that each will become the
creator of new narratives, of a new clinical sensi-
bility and art.

Questions and Answers

Question: Did you like the movie [of Awakenings]?
Sacks: I thought the movie was interesting. I
wasn't sure how a book, a thing made of lan-
guage, would translate to a film. When I first saw
the script I was absolutely bewildered — it con-
sisted entirely of images and things like voice-
over, which didn't convey anything to me. But by
and large, although there are things that I think
have been a little softened and sweetened here
and there, I think the film does give an honest,
tender picture of how things were in that summer
of 1969. I think it has a feeling of respect for the
patients and for the clinical process, and I do
like it.

Question: Do you use technology?
Sacks: A certain amount, everything from a re-
flex hammer to evoked potentials. I have nothing
against technology. I adore PET scans. I like
SQUIDS, you know, both sorts, all sorts. Among
some of my colleagues I sometimes see an identity
crisis. Neurologists wonder if they are replace-
able by a CAT scan. I think some of their diag-
nostic powers may be replaceable sooner or later
by a PET scan, but I think their judgment isn't
replaceable. No, I love technology. I want as much
as possible, but I want it to be combined with the
rest. And I like some words of Martin Buber: "We
must humanize technology before it dehumanizes
us." So I think we just must make technology part
of human life, and richly.

Question: How did you feel at the end of the sum-
mer of 1969, when things started to fall apart?
Sacks: Frantic, anxious, bewildered, guilty. I first
wondered if I was doing something wrong. Then
after a while I was persuaded that it was not I
who was doing something wrong, but that things
went wrong. Why they went wrong exercised me
for years and years, and when I started to read

Vol. 60 No. 2

NARRATIVE AND MEDICINE— SACKS

131

about the theory of chaos in the late 1970s and
the '80s, I suspected that it might be relevant.
What I saw toward the end of the summer of '69,
beyond the tragedies which were happening with
the patients — well, in a way nothing is beyond
tragedy — but in another realm what I saw was
the unpredictable. And science is based on predic-
tion. It was very threatening. It was epistemolog-
ically threatening to the patients, and to me, to
all of us, to see the unpredictable. One of the
things I like about some of the new theories,
chaos theory and catastrophe theory, is they
somehow make the unpredictable more under-
standable. And they bring new order out of chaos.
But one of the things I actually had to do was to
flee, which I did for three weeks, and to write, as
I did then, the first nine case histories oi Awak-
enings. So in some sense I went to narrative to try

to digest the experience, to make sense of it, and
to compose myself.

Question: Knowing the odds from your vantage
point now, would you do the whole thing over
again?

Sacks: Yes. I did hesitate for two years, despite
the very optimistic reports from 1967. Our pa-
tients had an infinitely more complex disease,
and they had been out of the world for decades. I
didn't know what would happen. The spirit of
doubt is stronger in me than the spirit of enthu-
siasm. I think the patients themselves at the
deepest level had no regrets, except to say that
they wished that L-dopa had been available ten or
twenty years before, when they would have had
less damage to their nervous systems, had lost
less life, had to make less accommodation, and
would have done better. I think that's the feeling.

Physicians Practicing Other
Occupations, Especially Literature

Jack Peter Green, M.S., M.D., Ph.D.
Abstract

Literature has been the favored nonmedical pursuit of physicians probably because the
practice of medicine is suffused with narratives, the patient's history being one. Arthur
Conan Doyle regarded medicine as a "grim romance," Somerset Maugham as an oppor-
tunity to see "life in the raw," and William Carlos Williams treated "the patient as a work
of art." These sentiments may be linked to humanistic medicine. At some medical schools,
literature is taught in the context of and integrated with medicine in an attempt to
enhance ethics and empathy which were explicitly expressed by some physician-writers.

Physicians have made contributions to many
fields other than medicine (1-4). St. Luke was a
physician and so was the man who became Pope
John XXI in 1276 (3). In Egypt 5000 years ago,
the physician Imhotep was deified (3). Deification
was not common, but Egyptians esteemed physi-
cians. In a papyrus, the powerful hawk-headed
sun-god Ra is contrasted with Sekhet the healer
(5). For Ra, after creating life, abandons the liv-
ing, whereas Sekhet lives among the sick, and he
mends, he heals, he cares. Imhotep was also a
scientist, engaged in architecture and astronomy
(3).

Other physicians made magnificent contribu-
tions to sciences other than medicine (3), includ-
ing Nicholas Copernicus (1473-1543), Carolus
Linnaeus (1707-1778), Luigi Galvani (1737-
1798), and Hermann von Helmholtz (1821-1894).
William Wollaston (1766-1828) discovered palla-
dium and rhodium, did research on the solar spec-
trum and on crystals, and invented the camera

An elaboration of an introduction to the lecture that Oliver
Sacks presented to the Lambda chapter of Alpha Omega Al-
pha on April 30, 1991 at the Mount Sinai Medical Center.
From the Department of Pharmacology, Mount Sinai School of
Medicine, New York, NY 10029. Address reprints requests to
Dr. Jack Peter Green, Department of Pharmacology, Box
1215, Mount Sinai School of Medicine, One Gustave L. Levy
Place, New York, NY 10029.

lucida. Peter Mark Roget (1779-1869) practice
medicine for fifty years and not only wrote th
Thesaurus, but invented a laboratory balance an
a calculating machine (6, 7). Thomas Youn
(1773-1829) did research on the eye (describin
astigmatism, the mechanism of vision, a theory c
color vision), revived the wave theory of light, o;
fered a theory of tides and elasticity, and, in a
extraordinary display of deductive ability, mad
the breakthrough in deciphering the Rosett
stone which was completed three years later b
Champollion (3). William Petty (1622-1687) cor
structed the first catamaran, did the first scier
tific surveying, served as professor of anatomy a
Oxford, helped found the Royal Society, pioneere
demographic analysis, and was the first economt
trician (3); one display of his originality in ecc
nomics is that he left a wealthy estate with
clause in his will that his debts were not to b
paid. The American physician Richard Gatlin
(1818-1903) invented agricultural tools as well a
the forerunner of the machine gun, which he ol
fered to the Union army, and it was accepted afte
the Civil War ended (4).

Physicians became politicians, too, and c
very different persuasions (8): Sun Yat Sen, th
first President of the Chinese Republic in 191^
Georges Clemenceau, Prime Minister of Franc
in 1917; Frangois Duvalier of Haiti; Che Guevar
of Cuba; Salvador Allende of Chile; and Howar

132

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 199

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

133

Dean, the present governor of Vermont. Five of
the 56 signers of the Declaration of Independence
were physicians (8).

The physician who provided the most perva-
sive and beneficent effect on politics was John
Locke (1632-1704). He practiced medicine much
of his life (9, 10). At Oxford, where Locke was
lecturer in Greek, in Rhetoric, and in Philosophy,
Robert Boyle provoked his interest in science. He
belonged to a group dedicated to reason and ex-
periment, which included Boyle, Thomas Willis,
and Richard Lower, a group that evolved into the
Royal Society (9). In his medical practice, he col-
laborated with the great Thomas Sydenham.
Locke's most memorable medical achievement
(10) was to insert a six inch silver tube into his
friend and patient. Lord Shaftesbury, to drain a
hydatid abscess; Shaftesbury wore the tube the
rest of his life. Locke's influence on political phi-
losophy cannot be exaggerated. In his ringing
proposal for the Constitution for Carolina, in
1669, he stated (10) that government rests on pop-
ular consent, that "All men are naturally in a
state of freedom, also of equality," and that "No
person whatsoever shall disturb, molest, or per-
secute another for his speculative opinions in re-
ligion or his way of worship."

It is believed (11) that some of the great
Greek philosophers — Empedocles, Pythagoras,
Democritus and Aristotle — studied medicine, as
did the Persian philosophers Rhazes and Avi-
cenna and the Spanish-Arabian Averroes. Albert
Schweitzer (1875-1965) wrote The Philosophy of
Civilization, as well as a biography of Bach and
articles on Goethe, all the while managing med-
ical care for the natives in Gabon, for which he
received the Nobel Peace Price, and gaining fame
as an organist. Other physicians contributed to
music (1, 3, 12, 13). After Alexander Borodin
(1833-1877) was graduated in medicine, he
worked on a research project and for relaxation
wrote Prince Igor; he also helped found a school of
medicine for women. Leopold Damrosch (1832-
1950) joined an orchestra soon after taking his
medical degree, and as a conductor, introduced
German opera at the Metropolitan Opera in New
York City. Sam Wong, the new assistant conduc-
tor of the New York Philharmonic, is an ophthal-
mologist. Painting has not been neglected.
Among contemporary physician-painters who
have exhibited are Joseph Wilder, a surgeon at
Mount Sinai Medical Center in New York; and
Roy Calne, professor of surgery at Cambridge
University, who does transplantations and con-
ducts research on transplantation (14). Two other

contemporary physicians became distinguished
in yet other pursuits: Roger Bannister, the first to
run the mile under four minutes, and Jonathan
Miller, the director of plays and operas.

As a physician is needed on every ship, phy-
sicians were aboard to minister to pirates and
buccaneers (15). Several rose to captain, one of
whom discovered Alexander Selkirk, the original
Robinson Crusoe, off the Chilean coast. Seagoing
physicians varied greatly in their temperaments.
At a time when cutting off ears for disobedience
was common, one physician dressed the wounds of
a prisoner despite emphatic objections by the cap-
tain. Not all were perceptive; one bought goat's
dung believing it was ambergris. Because the
chief surgeon was assigned to the flagship of the
fleet, a pirate crew voted their physician to be
chief surgeon in order to get rid of him. One phy-
sician was described as "a blustering braggadocio
rascal, full of windy schemes." Another was sen-
tenced to execution, but the sentence was never
carried out because he had helped the authorities
by betraying fellow prisoners, as he had earlier
betrayed a fellow physician:

A memory of roguery and laziness and knavery, you leave
behind you, young Doctor Wilson. Your colleague Comry in
the kindness of his heart fits you out with a clean shirt and
drawers so that you may appear presentably before your
pirate captain, you return his kindness by having him
forced aboard against his will. You refuse to dress the
wounded, and even your new pirate master tells you that
you are a double rogue to be with him a second time and
threatens to cut your ears off. All that saves your neck from
the rope is betrayal of your fellow prisoners' plans for free-
dom. Better for your reputation you had stayed ... at Cape
Montzerado and rotted there of your nasty leprous indispo-
sition (15).

They were men of their time and place, described

as men not so different from their present-day colleagues,
all kinds: the grave, the blustering, the whimsical, the
tranquil; the pompous, the cantankerous; the cocksure, the
lazy; the helpful, the kind; the roistering, the honest, and
the knavish — all kinds but the picayune and the squea-
mish. Their surroundings may have fostered crime and sin,
but no petty foibles; no paper rules and blanks to fill out,
nor quarantines and surveys and public health regulations;
only terrible foes, visible and invisible; terrible men, terri-
ble disease, terrible tempests, heat, cold, hunger and thirst;
terrible passion, greed, murder and savagery (15).

Many who became outstanding in other fields
had begun to study medicine but abandoned it (3,
6, 11). Included in this group are the scientists
Galileo, Humphrey Davy, Charles Darwin; the
composer Louis-Hector Berlioz; the painters
Jean-Frederic Bazille and Andre Breton. Others
who changed their course and became notable lit-
erary figures are Goethe, John Donne, Charles

134

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

Augustin Sainte-Beuve, Robert Southey, Samuel
Coleridge, Henrik Ibsen, James Joyce, Gertrude
Stein, Christopher Isherwood, and Georg Biich-
ner. Biichner (16) died in 1837 at the age of 23,
leaving behind three plays — each written in a few
weeks — "Brechtian" plays, a hundred years be-
fore Brecht, which have greatly influenced mod-
ern theater. He had to flee Germany, abandoning
medical studies, because of his political pam-
phlets urging human rights. In Zurich, where he
taught and did research on comparative anatomy,
he received a doctorate, but not in medicine (3, 6).

It is literature that has been the main other
calling of physicians, the subject of reviews and
compilations (1-3, 6, 7, 11, 12, 17-45). Biography
was not ignored. Harvey Cushing's biography of
William Osier won a Pulitzer prize (22). The
French surgeon Henri Mondor wrote biographies
of several French poets (26). But most of the work
was fiction and poetry. It has often been pointed
out that a linkage between medicine and the arts,
including literature, is represented in mythology
by Apollo, the god of poetry and god of the healing
art, and his son, Asklepius, on whom the mantle
of god of medicine fell. Theaters had inscribed
special seats for the priest of Asklepios, and the
theaters were located near the temples where
Asklepios effected cures, not other temples (46).
On the Acropolis in Athens, the theater and tem-
ple are connected by a ramp (46), and in Epidau-
rus, they face each other (46, 47). The propinqui-
ties have been suggested to be the reifications of
the belief in the contiguity of art and medicine
(47), or, at least, evidence that theater formed an
integral part of the worship of Asklepios (46), of-
fering "A healing word 'gainst many a foul dis-
ease / That all too long hath pierced with grievous
pains" (46).

Metaphors and Eponyms

Figures of speech suggest associations, if not
relations, between the arts and sciences, includ-
ing medicine. Scientific principles and discoveries
are common metaphors in the arts. Werner
Heisenberg's uncertainty principle has been
transmogrified to James Joyce's uncertainty prin-
ciple (48). It took little time for the new science of
chaos to become a consideration in literature (49).
The genetic code has been made musical (50).
Niels Bohr's complementarity principle is often
invoked to argue that two paradoxical views are
acceptable, even necessary. Metaphors and simi-
les from poetry are less often used in science. An
exhortation to biomedical scientists to learn

quantum mechanics promised introduction "to an
awesome achievement ... at once abstract, es-
thetic, rigorous and real, like Marianne Moore's
definition of a poem, 'an imaginary garden with
real toads in it' " (51).

Some eponymous diseases and syndromes de-
rive from literary characters (12). An advantage
of an eponym is that it can embrace succinctly
many manifestations of a disease that a more spe-
cific name may fail to include. For example, it has
been pointed out that Crohn's disease (named af-
ter the physician who first described the disease)
embraces pathology beyond the name regional il-
eitis (52). Munchausen's syndrome, from R. E.
Raspe's stories of the unbelievable, fabulous ad-
ventures of Baron Miinchhausen, is a patient's
presentation of feigned symptoms and often fab-
ricated signs of illness, convincing enough to lead
to costly and successive hospitalizations, exten-
sive tests, and invasive procedures that them-
selves produce morbidity (53, 54). The syndrome
is common enough to consider in differential di-
agnosis (55). Munchausen's syndrome by proxy is
the fabrication or induction of an illness in a child
by a parent (56). From time to time, patients dis-
missed as having Munchausen's syndrome were
later shown to have true disease, exemplified by
Lyme disease (57). Unlike malingerers, these pa-
tients have no apparent objective for faking. Like
the Baron, they travel widely, but their visits are
to hospitals. The saga of a man with the syndrome
was presented in verse (58):

He crashed through the door with a

horrible racket.
Two hundred and sixty pounds at least,
And covered with blood like a wounded
beast.

With gestures, he told us his chest was
stricken —

The coughed-up blood made the residents
quicken

To bring him assistance and ease his
anguish.

Residents don't let "an emergency"
languish.

Examined, his legs were red and swollen,
The large blue veins were easily rollen:
"Thrombophlebitis, then clot to the
lung"—

The residents knew this, although they
were young. . . .

His body was covered with many a scar.
He said from surgeons near and far.
The appendix went in County Cork,

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

135

A navel hernia in New York.
Once, he declared, in Portland, Maine,
A surgeon stripped out his saphenous
vein.

Surgical scars above one kidney
Came from an ectomy done in Sydney.
Another injury he wouldn't reveal us
Messed up his left internal malleolus. . . .

Munchausen's victims must be expected
To plague our lives unless deflected.
So be alert for this great nonesuchman —
Munchausen syndrome's flying
Dutchman.

Literary eponyms are sometimes strained or
literally inaccurate. Baron Miinchhausen's tales
were obviously incredible and meant to amuse,
whereas the lies of the patients are believable and
meant to deceive (12). The Raggedy Ann syn-
drome, named after a character in children's
books created by J. Gruell, is a condition of
chronic lassitude, which does not describe the
lively Ann but rather the flaccid doll that was
made to publicize the book (12). The Pickwickian
syndrome, extreme obesity associated with som-
nolence and hypoventilation, should be called the
Fat Boy Joe syndrome, since Joe had it, not Pick-
wick, in Dickens' novel (12). Another eponym
that distorts its literary root is the Oedipus com-
plex, which only a convoluted extrapolation could
relate to Sophocles' story of Oedipus Rex, who did
not know until too late that the man he killed was
his father or that the woman he married was his
mother. Similarly, referring to exhibitionism as
the Lady Godiva syndrome misrepresents the leg-
endary Lady, whose body was hidden by her long
hair in her gallop through Coventry, the purpose
of which was not exhibitionism but to shock her
husband into reducing taxes (12).

These literary eponyms are metaphors,
which, John Locke insisted, are "wholly to be
avoided" because "they insinuate wrong ideas"
(59). Metaphors, like analogies, are dangerous if
used in philosophy or in serious discussions of pol-
icy because they are extrapolations of one circum-
stance or thing to another that is intrinsically
different. And they are dangerous if accepted as
actualities rather than as figures of speech (60).
Metaphors can offer vivid descriptions and stim-
ulate useful associations. Medical literary ep-
onyms, however unfaithful they may be to their
literary sources, do not mislead, since all physi-
cians agree on what they connote in clinical con-
text.

Just as literature has enriched medical lan-

guage, physicians have given their names to the
English language. The verb bowdlerize comes
from the physician Thomas Bowdler (1754-1825),
who published expurgated versions of Shake-
speare, the Old Testament and Gibbon's History of
the Decline and Fall of the Roman Empire, ex-
punging words and expressions that could not
"with propriety be read aloud in a family" (3). It
has been pointed out that if these omissions had
not been made, young people would not have been
permitted to read the books (3).

Very unbowdlerized is the work of another
physician, Francois Rabelais (ca. 1494-1553),
who gave us the word rabelaisian, derived from
the bawdy, robust, scathing humor of his novels
(61, 62). His Gargantua and Pantagruel are said
to be drugs that would provide understanding and
therapy (61, 62). He received a medical degree
from Montpelier after becoming a priest. He
moved to Lyons, where he taught anatomy, doing
human dissections against church policy. Along
with ribald novels he wrote medical articles and
the first Latin translation of Hippocrates' apho-
risms. His work embraced all that was known of
Renaissance medicine (61). His fiction was con-
demned for mockery of religious practices, but he
survived, probably because he was a friend of a
cardinal, other churchmen, and politicians, and
he was acquainted with the king. He is regarded
as one of the greatest of all satirists and comic
geniuses.

Physician-Writers and
the Despotisms

Celine. Unlike Rabelais, some other physi-
cian-writers joined the despots or were despoiled
by them. Louis-Ferdinand Celine (born Des-
touches) (1894—1961) practiced medicine almost
all his life while writing, according to one critic,
some of the greatest fiction in history (63). He was
described as a kind and gentle physician, all the
while writing novels that create a "garden of hu-
man disintegration" (64), searing and savage.
Physicians are protagonists in some of the novels,
which offer foul, gripping images of disease. He
urged alliance with the Nazis and wrote violently
anti-Semitic and racist pamphlets for them dur-
ing the occupation. After the war, he fled to Ger-
many, then to Denmark, where he was denounced
as a collaborator in 1946 and imprisoned for 14
months (65, 66). The French condemned him in
1950 in absentia to a year in prison and a fine and
declared him a "national disgrace" (65, 66). He

136

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

was then given amnesty (66) and returned in
1951 to practice medicine and write. His house in
Meudon, a Paris suburb, is being proposed as a
listed historical monument by the French cul-
tural ministry. Meudon provides a convergence of
two physician-writers; 400 years earlier, Rabelais
served as curate there shortly before he died (67).

Benn. Not well known in America is another
physician-writer who was a Nazi for a short time,
the German Gottfried Benn (1886-1956), who
practiced both occupations almost to the end of
his life (26, 68, 69). He wrote poems, plays, es-
says, short stories, and an oratorio that was set to
music by Hindemith. As a medical student, he
also studied philosophy and art while writing po-
ems and short stories. He won a prize in 1911
from the University of Berlin for a study of epi-
lepsy. His artistic goal was a synthesis of "scien-
tifically exact description with modern expres-
sion" (68). The goal was attained early (1912)
with the publication oi Morgue and Other Poems.
Morgue contains grisly descriptions of autopsies.
The volume propelled him into fame. As a physi-
cian assigned to the German military govern-
ment in Brussels in 1915, he participated in the
trial of the English nurse Edith Cavell and signed
the death certificate after she was shot as a spy.
He practiced in Berlin from 1917 until 1935, spe-
cializing in dermatology and venereal diseases.
His life (68) was chaotic and not benign. Soon
after he was admitted to the poetry section of the
Prussian Academy of Arts, Hitler came to power,
in 1933. Heinrich Mann, president of the section,
and Kathe Kollwitz, the artist, voiced opposition
to the regime, warning of a relapse into barba-
rism. Benn replaced Mann as president, and did
not protest the persecution of his Jewish col-
leagues. Rather, from 1933 to 1934, he supported
the Nazis in essays and speeches, praising Hitler,
demanding that intellectual freedom be sacrificed
for the good of the state and for the future of the
German people. Although his father and grand-
father were Protestant ministers, he was de-
nounced as a "pure-blooded Jew," and his certifi-
cation as a physician was withdrawn. He had to
prove his Aryan extraction. In 1934, he wrote a
defense of modern art, which caused him again to
be condemned, this time as a decadent cultural
Bolshevist. He continued to avow loyalty to the
Nazis until their crushing of an alleged plot
changed his mind. He rejoined the army, which
served as a refuge from Nazis, and wrote poems
and essays to the end of 1944, criticizing the Nazi
movement and both the Marxists and Nazis for
extolling the collective rather than the individ-

ual. The Nazis prohibited his literary work from
being published, as did the Allies during the oc-
cupation. In 1946, his attempt to publish the work
he had accumulated since 1936 was blocked in
Germany by Alfred Doblin, who in 1933 had been
one of the ousted members of the poetry group.
The Swiss published it in 1945, later the Ger-
mans. Benn's favorite biblical quotation, from
Jeremiah, "God has walled me in, so that I cannot
escape, and has laid heavy chains about me," is
more appropriately applied to some of his contem-
poraries than to him. For him, the quotation has
lost the pathos it may have had (if self-pity is
granted pathos), as he was awarded numerous
honors, including the Biichner prize, and he is
now regarded as the most important German poet
since Rainer Maria Rilke.

Doblin and Levi. Alfred Doblin (1878-1957),
who temporarily thwarted the publication of
Benn's later work, was a physician-writer. One of
his somewhat expressionistic novels, Alexander-
platz, Berlin, stems from his experiences in a
working class neighborhood (26, 70). A Jew, he
left Berlin in 1933, came to America, and then
settled in France after the war. He stopped med-
ical practice after leaving Germany. His novels
received much praise from Bertolt Brecht, not an
easy achievement (70).

The Italian Jew Carlo Levi (1902-1975) did
nothing medical with his degree, instead becom-
ing an acclaimed painter and writer (26). Because
he was an anti-Fascist leader, he was exiled in
1935 to Lucania in the south of Italy. His book
Christ Stopped at Eholi is based on his experi-
ences there. He also wrote essays.

Wolf. Friedrich Wolf (1888-1953), a Jewish
physician-writer, studied medicine at Heidelberg
while also studying sculpture and painting (71).
He worked as a model for books on anatomy, dis-
playing muscles that were well defined probably
because he was active in track and field. He be-
gan to write stories and plays while serving as an
Army doctor in World War I. He achieved fame as
a dramatist in 1924, and for the rest of his life
wrote plays, novels, stories, a libretto, movie and
radio scripts, and Soviet propaganda. He stopped
medical practice in 1933 when he fled Germany
because of his political activities, eventually go-
ing to Paris. He had joined the Communist party
in 1928, and worked and made speeches for the
party. After the French released him from prison
in 1940, he moved to Moscow, and in 1945 he
settled in East Berlin. He was an East German
diplomat to Poland and helped set the German-
Polish border. His output was copious (71). His

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

137

best-known work is the play Professor Marnlock
(72), written in 1933 and produced in many coun-
tries besides Germany, including America. His
son Markus was the leader of the very effective
East German spy ring during the Cold War.

Bulgakov. A physician-writer who was al-
most destroyed by a totalitarian state was
Mikhail Bulgakov (1891-1940), described as one
of the greatest writers of the twentieth century
(73, 74). He was graduated with distinction from
Kiev University medical school in 1916, and four
years later abandoned medicine to begin a career
as a writer (73, 74). For the first year and a half of
practice, without benefit of an internship, Bulga-
kov was assigned to provide medical care in a
wilderness where wolves had to be kept at bay
with a pistol on returns from night calls (75). The
nine stories of his experiences there, real and
imagined, have been collected (75).

One, Morphine, is presented as the diary of a
physician who becomes dependent on morphine.
It reads like numerous case histories, the horror
and pity enhanced by Bulgakov's imagination.
From the certitude that he will not become depen-
dent, the physician declines inexorably to "surely
mild habituation is not the same as becoming an
addict?" then to injecting over 100 mg twice a day
of "those life-giving crystals." Fear of exposure,
the ruses used to get the drug, and the auditory
and visual hallucination are presented: "At
dusk — always my worst time — I clearly heard a
voice in my room, monotonous and threatening,
repeating my name. ... It stopped as soon as I
injected myself." And "the old woman was not
running but actually flying, without touching the
ground. This was bad enough; but what made me
scream aloud was the fact that she was holding a
pitchfork in both hands. ... I fell on to one knee,
holding out my hands to shield myself from the
sight, then I turned and ran, stumbling, for home
and safety, praying that my heart would not give
out before I could reach my warm room . . . and
take some morphine." All other interests are
eroded. When asked by his mistress, "What can
bring you back to life? Perhaps . . . that opera
singer?" he answers, "I've got over her. ... I have
morphine instead." He tries cocaine:

after giving my puncture-riddled thigh a careless smear of
iodine, I dig the needle into the skin. Far from feeling any
pain, I have a foretaste of the euphoria which will overtake
me in a moment. And here it comes. I am aware of its onset,
for . . . the faint, muffled snatches of music sound like an-
gelic voices, and the harsh bass chords wheezing from the
bellows ring out like a celestial choir. But now comes the
moment when, by some mysterious law that is not to be
found in any book on pharmacology, the cocaine inside me

turns into something different. I know what it is: it is a
mixture of my blood and the devil himself. The sound of . . .
music falters . . . while the sunset growls restlessly and
burns my entrails. This feeling comes over me several times
in the course of the evening, until I realize that I have
poisoned myself. My heart begins to beat so hard that I can
feel it thumping when I put my hands to my temples. . . .
Then my whole being sinks into the abyss. . . . I . . . who
became addicted to morphine . . . warn anyone who may
suffer the same fate not to attempt to replace morphine
with cocaine. Cocaine is a most foul and insidious poison . . .
today I am half dead.

Self-loathing increases and finally, suicide.

In 1928 Bulgakov had three plays running in
Moscow, all of which were removed by censors
whose objection to the content of the plays was
exacerbated by their popularity (74). Of the 19
plays he wrote during his lifetime, eight were per-
formed and 11 published in the Soviet Union (76).
Maybe subscribing to Chekhov's belief that
"great writers and artists should engage in poli-
tics only to the extent needed to defend them-
selves from politics," he wrote to Stalin in 1930,
stating that 298 of the 300 reviews of his work
were abusive and that the reviewers were correct
in asserting that his work cannot exist in the
USSR because "It is my duty as a writer to fight
against censorship, whatever its forms and under
whatever government it exists, and to call for
freedom of the press. ... I appeal to the humanity
of the Soviet government, and beg that I, a writer
who cannot be useful in his homeland, be mag-
nanimously set free and allowed to leave" (73-

76) . Stalin answered Bulgakov by phone, refusing
emigration but arranging employment at the
Moscow Art Theatre, where Bulgakov served as
literary consultant and assistant director. While
dramatizing the work of others (War and Peace,
Don Quixote) and writing librettos and movie
scripts (74), he continued to write his own plays,
novels, short stories, and a biography of Moliere.
The biography was not published, and a play
about Moliere, probably because of the enthusias-
tic response from the audience, was closed after
seven performances on attack from Pravda (73,

77) . The play depicted Moliere's harassment near
the end of his life by the sycophants surrounding
Louis XIV, a circumstance, evident to everyone,
that paralleled Bulgakov's treatment. In the six-
ties, some of Bulgakov's work was published, not
surprisingly absent the novel Heart of a Dog (11).
It is the story of a dog who is transformed by
transplants into a repugnant, Engels-speaking
man and becomes the government official as-
signed to purge the city of cats.

Bulgakov's work, like Chekhov's, showed ha-

138

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

tred of oppression, and Bulgakov suffered for it.
He was not corrupted or destroyed by the humil-
iations he endured — comparatively trivial ones
like confiscation of his diaries and searches of his
home — but he was ravaged by unremitting, vitu-
perative opprobrium and vilification by the ideo-
logical zealots of a brutal political system. Bulga-
kov's torment by the Soviets is reminiscent of the
treatment of Alexander Pushkin by the tsars,
who censored his work, much of which still re-
mains unpublished 150 years after his death.

Three years before dying, Bulgakov wrote to
a friend, "Some well-wishers have chosen a rather
odd manner of consoling me. I have heard again
and again suspiciously unctuous voices assuring
me, 'No matter, after your death everything will
be published' " (74). The well-wishers were right.
Not only are his works being published in Moscow
as in the west, but the government gave him of-
ficial honors on the hundredth year of his birth
(78). His apartment in Moscow has become
shrinelike, a place where people gather to read
and discuss his works, covering the doors, walls
and ceilings with tributes and graffiti, including
"No to the Devil" (78), a reference to one of his
great novels. The Master and Margarita (73), im-
plicitly addressed in the novel to Bulgakov's gov-
ernment.

Some Medical Truants

Like Wolf, Doblin, Levi, and Bulgakov, many
other physicians whose writings were successful
abandoned medicine, joining the "medical tru-
ants" (1), defined as physicians who left medicine
to become distinguished in other fields. Opposite
odysseys were those of T. L. Beddoes, who left lit-
erature for medicine (18), and James Clark, who
took up medicine after he failed as a writer, be-
coming Queen Victoria's physician (6).

Many literary critics would probably agree
that among the truants, those who had the great-
est effects on literature were Oliver Goldsmith
(1730-1774), Friedrich von Schiller (1759-1805),
John Keats (1795-1821), and Anton Chekhov
(1860-1904).

Goldsmith. Goldsmith (6) began writing and
selling ballads while a student at Trinity College
in Dublin. He was generous. He impulsively gave
his blankets to a poor family in Dublin and slept
among the feathers of his bed to keep warm.
When he went to take a boat in Cork, he met a
woman with eight children and gave her all his
money; his brother paid for his return to Dublin.

His uncle gave him money to study law in Lon-
don, which he gambled away before leaving Dub-
lin. He studied medicine in Edinburgh, where
Irish students rescued him from debt. He tried to
practice medicine in London, and a friend report-
edly advised him "to give up medicine on the
grounds that if he were resolved to kill he should
concentrate on his enemies." The Vicar of Wake-
field and She Stoops to Conquer are considered
classics, and he wrote poems as well, one of which.
The Deserted Village, has been called "the sweet-
est poem in the English language" (11). He died,
as he had lived, much loved and in debt. He was
described by Oliver St. John Gogarty as "the gen-
tle Irish Virgil."

Schiller and Keats. The German Schiller was
forced by a duke to study medicine and to serve as
an army surgeon before beginning his writings,
which excited the wrath of the duke, compelling
him to flee and to become, along with his friend
Goethe, one of the founders of modern German
literature (6). His Ode to Joy, used for the finale of
Beethoven's Ninth Symphony, is one of his offer-
ings. The Englishman Keats left the practice of
medicine in 1806 at the age of 21 to write, in the
remaining four years of his life, some of the great-
est lyric poetry in the English language (21, 79-
81). One encomium is that he is "the greatest of
Shakespeare's successors" (18).

Chekhov. Chekhov certainly ranks among
the greatest of physician-writers. Besides plays,
he wrote short stories, novels, and volumes of let-
ters. While studying medicine in Moscow on
scholarship, he earned money to support his fam-
ily by writing sketches. He received his degree in
1884. A monograph resulted from his three-
month visit to the penal colony at Sakhalin, an
island off Siberia, when he was 30 years old, after
he had become known as a writer. In this, his only
nonfiction work, he includes statistics on the is-
land, based on interviews of all its inhabitants
(82). Writing this took three years, and elicited
his complaint of being "forced for the sake of a
single mangy line or other to rummage among
papers for a full hour" (82). He noted that one of
the causes of death listed in records on the island
was "lack of development towards life" (82), her-
alding the senseless diagnoses of dissidents made
by some Soviet psychiatrists. After returning to
Moscow, he arranged that thousands of books be
sent to the schools on Sakhalin Island. One rea-
son for visiting Sakhalin, he said in a letter
(March 9, 1890), was, "I want to write one hun-
dred to two hundred pages and thereby pay off
some of my debt to medicine, toward which, as

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

139

you know, I have behaved like a pig"; another
reason that he gave after submission of the manu-
script (January 2, 1894) was, "Medicine cannot
now accuse me of infidelity: I have rendered just
tribute to learning and to that which the old writ-
ers used to call pedantry" (83).

He sporadically practiced medicine, notably
in 1892 when he worked to contain a cholera ep-
idemic. His major commitment was to literature,
as is clear from the progression of letters (83),
some of them to Suvorin, a critic and — until Su-
vorin reviled the supporters of Dreyfus — a friend.
One dated January 17, 1887: "Besides medicine,
my wife, I have also literature — my mistress, but
I do not mention her — those living in sin will per-
ish sinfully." On February 11, 1893: "Medicine is
my lawful wife, literature my illegitimate spouse.
Of course, they interfere with each other, but not
so much as to exclude each other." The wife did
not triumph. On November 11, 1893: "I've become
so devoted to it [literature] that I've begun to de-
spise medicine." On January 21, 1895: "I am
obliged to literature for the happiest days of my
life and for what I am chiefly drawn to."

Chekhov wrote to a friend, "My Holy of Ho-
lies is the human body, health, intelligence, tal-
ent, inspiration, love and the most absolute free-
dom— freedom from violence and falsehood, in
whatever form these may be expressed" (7).
Health was the only one of the Holies he lacked.
Tuberculosis and its hemorrhages assailed him
most of his life, killing him as it had Keats. De-
spite a short life, his collected writings comprise
20 volumes. Maxim Gorky said, "In Chekhov's
presence, everyone wished to be simple and truth-
ful, to become himself (7). He has been described
as "a secular saint, the most attractive personal-
ity in the whole history of literature" (7).

Schnitzler. Arthur Schnitzler (1862-1931)
practiced laryngology in Vienna, like his very ac-
complished father, while writing plays, poetry
and novels (22). Physicians appear in one-third of
his work. He dropped medicine before he turned
forty to write full time. He is best known in Amer-
ica for La Ronde, a witty, clever play that was
made into a movie by the French. The play was
attacked as "pure smut" by the Society for the
Suppression of Vice, and its staging in New York
was prevented in 1923 (84).

His novel The Road into the Open has been
described as a neglected masterpiece, "a book
that the 20th century has misplaced" (84), and
Schnitzler was described as "the doctor who left to
literature his diagnosis of a society in crisis" (84).
His work reflects his interest in psychiatry, caus-

ing Freud to describe him as his "double" and to
write "Whenever I get deeply absorbed in your
beautiful creations I invariably seem to find be-
neath their poetic surface the very presupposi-
tions, interests and conclusions which I know to
be my own" (84). After Hitler entered Vienna in
1938, the Burgtheater announced, "The day has
come on which the stage of the Burgtheater is to
be cleansed of the Judenschmutz, which was
dumped there by Schnitzler and his consorts"

(84) . In 1941, at a celebration of Schnitzler's
birthday ten years after his death, a Viennese
emigre called Schnitzler one of "the voices that
sound above the chaos, because their ring is so
pure, because their prophecy is so true, because
their humanity is so great" (84). Schnitzler's fa-
ther had declared that "The religion of the doctor
is humanity . . . the love of mankind" (84).

Some Irish, English, and Other Truants. Ire-
land has made an inordinate contribution to the
list of physician-writers. One Irishman, Charles
Lever (1806-1872), like his compatriot Gold-
smith, composed ballads while at Trinity College,
and he also sang them on the street for money. He
gave up medicine in 1842, eleven years after re-
ceiving his medical degree. He wrote 30 novels in
35 years, becoming one of the most popular nov-
elists of his time. One of his picaresque adven-
tures occurred in Canada, where, after being ad-
mitted to an Indian tribe, he decided to resign. He
was sentenced to death by tomahawk, but two
Indians helped him to escape (6).

Somerset Maugham (1874-1965) embraced
full-time literature right after his internship,
writing novels, plays, short stories, and essays

(85) . His first novel, Liza of Lambeth, is based on
his experiences in a London slum during medical
training. Of Human Bondage is his novel about a
medical student burdened by a clubfoot and, far
worse, an obsession with as loathsome a woman
as ever appeared in fiction. He was financially
successful and the urbane sophisticate. Near the
end of his long life, he became irrational. On a
railway platform, a woman approached
Maugham's companion to urge, "You should be
gentle with this nice old man. He thinks he's
Somerset Maugham" (6).

The British poet laureate Robert Bridges
(1844—1930), who was also a playwright, gave up
medicine at 40. A. J. Cronin (1896-1981) gave up
a prosperous London practice to express his med-
ical knowledge in best-selling novels. Similarly,
Pio Baroja y Nessi (1879-1956) was a physician
who became the most popular Spanish novelist of
the 20th century (39). The French surgeon

140

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

Georges Duhamel (1884—1966) wrote novels and
plays. One of the few American truants was the
novelist Walker Percy (1916-1990), who gave up
medicine when he became ill as an intern.

George Crabbe (1754-1832) failed as a sur-
geon and, though not a very successful poet, pro-
vided a description of physicians in verse (11):
Men who suppress their feelings, but who
feel

The painful symptoms they delight to
heal;

Patient in all their trials they sustain
The starts of passion, the reproach of
pain,

With hearts affected, but with looks
serene;

Intent they wait through all the solemn
scene

Glad if a hope should rise from nature's
strife

To aid their skill and save the lingering
life.

Arthur Conan Doyle. Scots are a major group
of physician-writers, comprising one-third in one
compendium (6). Among them, or perhaps among
all writers, no one wrote more popular fiction
than Arthur Conan Doyle (1859-1930). His cre-
ation, Sherlock Holmes, has obscured other con-
tributions of Doyle that have been recalled and
praised (86-90). Doyle's grandfather emigrated
from Ireland to Edinburgh and achieved fame as
a caricaturist, as did one of Doyle's uncles, who
did cartoons for Punch, designed its first cover,
and did illustrations for Dickens and Thackeray
(6). Doyle went to a Jesuit school and, at his
mother's urging, studied medicine at the Univer-
sity of Edinburgh. Doyle's laments about his med-
ical education (91) over a hundred years ago are
echoed by students today, "one long dreary grind
of ... a whole list of compulsory subjects, many of
which have a very indirect bearing upon the art of
curing. The whole system of teaching, as I look
back upon it, seems far too oblique and not nearly
practical enough." And "I think that our educa-
tional tendency . . . was to expend undue atten-
tion upon rare diseases, and to take the common
ones for granted. Many men who were quite at
home with strange pathological lesions found
themselves in practice without ever having seen a
case of scarlatina or measles." A man of "Her-
culean proportions" (91), he excelled in many
sports, including boxing (92). As a medical stu-
dent, he spent seven months on an Arctic whaling
vessel as a ship's surgeon. One of the crew with
whom Doyle had boxed said Doyle is "the best

surgeon we've had. He blacked my eye" (88). His
prowess as a harpooner prompted an offer to dou-
ble his salary if he would sail again (91). While
doing well in his medical studies, he sold his fic-
tion, and also worked as a physician's assistant to
help support himself (88, 92).

After graduation in 1881, he served four
months as a medical officer on a ship sailing to
West African ports, and soon after returning, ac-
cepted a partnership in the practice of a friend
from medical school, George Budd. Budd's prac-
tice was unusual. Each patient was given a ticket
with a number designating his or her turn to see
Budd; for a fee, the patient could move to the head
of the line (88). Except for those patients in a
hurry, consultations were free, and in return the
patients agreed to pay for prescriptions that Budd
dispensed and his wife bottled. The overflow oi
patients required a partner (91, 93). Budd became
a character in Doyle's The Last World, which was
made into a movie (91). Later, Doyle described hij
last meeting with Budd, quoting him: "I've taker
to the eye, my boy. There's a fortune in the eye. A
man grudges a half-crown to cure his chest or his
throat, but he'd spend his last dollar over his eye
There's money in ears, but the eye is a gold mine!'
(93). Doyle left, but with no acrimony, for Budc
provided Doyle one pound a week while he estab-
lished his own practice (93). In a town healthiei
than its neighbors and one with a surfeit of phy-
sicians (90), Doyle's practice was moderately suc-
cessful. During this period, in 1886, he wrote his
first Sherlock Holmes story, the Study in Scarlet
Rejected by three respected publishers, it was ac-
cepted by a house specializing in sensational lit-
erature that paid Doyle 25 pounds for rights thai
included the complete copyright, which eventu-
ally embraced movie rights (93).

Doyle became interested in ophthalmolog)
and worked in an eye and ear infirmary. He weni
to Vienna for two months to learn ophthalmology
and then opened a practice in London to do refrac-
tions and retinoscopy (88, 90). In eight weeks, nc
patients arrived. He spent his time writing anc
found a prominent literary agent. In 1892, he
gave up practice to make a living as a full-time
writer. His only subsequent medical experience
was to serve four months as a physician in the
Anglo-Boer War, for which he was knighted ir
1902 (89, 93). The common belief that Doyle
wrote fiction because he failed as a physician has
been ascribed to the Bellman's fallacy, the repe-
tition of statements from secondary sources with-
out examining original, primary sources (12). It is
certainly likely that even if Doyle's practice had

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

141

been highly successful, he would have continued
to write fiction, as he had been doing all his life.

The name Holmes was chosen because of
Doyle's admiration for Oliver Wendell Holmes,
about whom Doyle said, "Never have I so known
and loved a man whom I had never seen. It was
one of the ambitions of my lifetime to look upon
his face" (93). He visited Holmes' grave in 1922-
1923. While in New York, Doyle lived at 310
West 75th Street (94). The character Holmes was
inspired partly by Doyle's study of the Edgar Al-
lan Poe stories about C. Auguste Dupin, the am-
ateur detective who worked solely by deduction.
Holmes mostly evolved from Doyle's experiences
with Joseph Bell, his teacher at medical school.
Doyle wrote Bell in 1892, "It is most certainly to
you that I owe Sherlock Holmes. ... I do not think
that his analytical work is in the least an exag-
geration of some effects which I have seen you
produce in the outpatient ward. Round the centre
of deduction and inference and observation which
I have heard you inculcate I have tried to build up
a man who pushed the thing as far as it would go"
(95). The resemblance between Holmes and Bell
was so clear that Robert Louis Stevenson wrote
Doyle from Samoa, asking, "Only one thing trou-
bles me; can this be my old friend Joe Bell?" (96).

Joseph Bell descended from a family of prom-
inent physicians in Edinburgh. Bell wrote two
successful books on surgery, edited the Edin-
burgh Medical Journal for 23 years, and was re-
vered by patients and students. He astonished ev-
eryone with his diagnostic acumen. Bell exhorted
his students to "use your eyes, use your ears, use
your brain, your bump of perception, and use your
powers of deduction" (97). At the same time and
independently, another great physician, William
Osier, was urging his students to "Use your five
senses. . . . Learn to see, learn to hear, learn to
smell . . . for the whole art of medicine is in ob-
servation" (98). A successful diagnostician needs
"an imagination capable of weaving a theory or
piecing together a broken chain or unraveling a
tangled clue" wrote Bell in a foreword to one of
the Sherlock Holmes books (97). Parallels be-
tween statements by Holmes and by Bell have
been collated (88, 97).

In developing Sherlock Holmes, Doyle wrote,
"I thought of my old teacher, Joe Bell ... of his
eerie trick of spotting details." Revealing how
Doyle emulated Bell and why Bell regarded Doyle
as one of the best students he ever had and why
he had selected Doyle to be his outpatient clerk is
Bell's description of Doyle: "He was exceedingly
interested in anything connected with diagnosis,
and was never tired of trying to discover all those

little details which one looks for" (95, 96). Asked
to describe the details that provide salient diag-
nostic information. Bell offered an anecdote:

A man walked into the room where I was instructing the
students, and his case seemed to be a very simple one. I was
talking about what was wrong with him. "Of course, gen-
tlemen," I happened to say, "he has been a soldier in a
Highland regiment, and probably a bandsman." I pointed
out the swagger in his walk, suggestive of the piper; while
his shortness told me that if he had been a soldier, it was
probable as a bandsman. Well, the man . . . said he had
never been in the army in his life. . . . Being absolutely
certain I was right, and seeing that something was up, I did
a pretty cool thing. I told two of the . . . dressers, to remove
the man to a side room, and to detain him till I came. I went
and had him stripped . . . under the left breast I instantly
detected a little blue "D" branded on his skin. He was a
deserter. That was how they used to mark them in the
Crimean days, and later, although it is not permitted now.
Of course, the reason of his evasion was at once clear (96).

Bell was not averse to telling anecdotes of his
diagnostic gaffes: "You are a bandsman?" "Aye,"
replied the sick man. Dr. Bell cockily turned to
his students, "You see, gentlemen, I am right.
This man has a paralysis of his cheek muscles,
the result of too much blowing at band instru-
ments." And turning again to the patient, "What
instrument do you play, my man?" "The big
drum," came the reply (99).

Bell initially resented the publicity following
Doyle's revelation of the inspiration for Holmes,
writing to a friend, "Why bother yourself about
the cataract of drivel for which Conan Doyle is
responsible? I am sure that he never imagined
such a heap of rubbish would fall on my devoted
head in consequence of his stories" (99). Later, in
addition to writing a foreword to one of the
Holmes books, he sent Doyle suggestions for plots
(97). Bell himself had a liking for criminal work;
he collaborated with Henry Littlejohn, the police
doctor in Edinburgh and later a professor of med-
ical jurisprudence, solving major crimes (96).

Bell described a pervasive yield of the
Holmes stories: "I believe they have inculcated in
the general public a new source of interest. They
make many a fellow . . . think that, after all,
there may be much more in life if he keeps his
eyes open. . . . There is a problem, a whole game
of chess, in many a little street incident or trifling
occurrence if one once learns the moves" (96). A
century later it is independently stated that "ed-
ucation in the arts . . . provides an indispensable
means of helping us develop our capacity to
see. . . . The world's objects and situations, its ev-
eryday events, are replete with meaning waiting
to be seen and heard by the watchful eye and the
attentive ear" (100).

142

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

Sherlock Holmes became the greatest legend-
ary figure of our time (90, 96), a mythic creature
metamorphosed to a living person, provoking
Sherlockitis, Sherlockidolatry [shortened to Sher-
lockolatry (101)], and in South America, the word
Sherlock-holmitos, which is a pointless but im-
pressive deduction (12). Holmes stimulates spec-
ulation of his true Baker Street address and the
formation of hundreds of societies dedicated to
him. The Baker Street Irregulars proclaim Wat-
son as the true author of the Holmes stories, re-
ferred to as "The Sacred Writings," Doyle as his
literary agent (102). Another suggestion is that
Watson really solved the cases, but credited
Holmes because of his innate modesty, his gentle-
manly tendency to self-deprecation, and the fear
of penalties that would be exacted on a physician
who advertises (103). Publications abound about
Holmes — the Baker Street Journal, Baker Street
Misc., The Sherlock Holmes Gazette, and at least
eight biographies. Holmes was awarded an hon-
orary diploma from Colorado State University,
and his picture appears on postage stamps (90).
There are arguments about which university
Holmes attended (6). A genealogy of Holmes con-
cludes that "Nero Wolfe, purportedly a fictional
creation of Rex Stout, was no doubt the illegiti-
mate son" of Holmes (104). He has been the sub-
ject of over 300 movies, over 75 television shows,
over 70 plays, over 30 radio series, the subject of
musicals and ballet (90). Parodies of the Holmes
story were written from the beginning. James
Barrie, Doyle's friend and patient and author of
Peter Pan and Wendy, wrote a parody to amuse
Doyle (6). The physician Logan Clendening gave
an account of Holmes' arrival in heaven, where
Jehovah requests him to find the missing Adam
and Eve, who had been major attractions there.
Holmes succeeds. Jehovah wonders how Holmes
so quickly identified them in a crowd, and Holmes
replies, "Elementary, my dear God, they have no
navels" (92). Rex Stout concluded that Watson
was really a woman (102). William Gillette, the
American actor, asked Doyle for permission to
marry Holmes (102). One may anticipate the sug-
gestion that Holmes predicted the Actors Studio
method when he stated, "The best way of success-
fully acting a part is to be it" (97). And a prospec-
tive father may hope that a demand for equality
of the sexes is not prophetic in the description of a
woman who "insisted that her husband be hand-
cuffed to her during her entire labor, which lasted
eight hours. With the twisting during severe
pains the iron had fairly eaten into the bone of the
man's arm. She showed no remorse. 'He's got to
take his share as well as me' " (89). On Robert

Louis Stevenson, Holmes effected unusual cures,
as Stevenson wrote Doyle from Samoa, struck by
the "ingenious" adventures of Holmes: "That is
the class of literature I like when I have the tooth-
ache. As a matter of fact, it was pleurisy I was
enjoying when I took the volume up; and it will
interest you as a medical man to know that the
cure was for the moment effectual" (96).

Holmes' logic can be used to illustrate critical
reasoning (105). His method of deduction is the
same as that practiced by physicians in differen-
tial diagnosis (44), using erudition, rigor, and
logic, precisely identifying all pertinent findings
to diagnose the disease while eliminating others
by precisely noting the absence of their signs and
symptoms. A differential diagnosis was applied to
the disease that Holmes feigned in order to catch
a murderer in The Adventures of the Dying Detec-
tive (101). Holmes described it as a "coolie disease
from Sumatra — a thing the Dutch know more
about than we." The clever differential diagnosis
(101) rules out metazoan, protozoan, spirochetal,
and fungal infections, concluding it was tsutsu-
gamishi disease, also called Sumatra mite fever
and inundation fever, first described by the Jap-
anese and encountered by the Dutch in Sumatra,
where workers in rice paddies were bitten by the
mites when the paddies overflowed.

Amusing diagnoses have been proposed to ac-
count for apparently contradictory statements by
Dr. Watson about his war wound (106). In one
Holmes story, Watson mentions having been
struck by a bullet in the shoulder, and in a sub-
sequent story, mentions that a bullet passed
through his leg. One proffered explanation for the
paradox is that the bullet in the shoulder caused,
on occasion, diminished cerebral blood flow, lead-
ing Watson to forget the correct site of his injury.
Another is that Watson was bending over a pa-
tient when hit in the shoulder with the bullet,
which then ricocheted under the skin, traveling
to his leg. A third explanation is that the bullet
ricocheted after striking a bone in the shoulder,
leaving the body to enter the leg. Other ingenious
explanations to support a one-bullet theory have
been offered. "Briefly, he was squatting over the
edge of a precipice, of which there are many in
Afghanistan, to answer the call of nature; and he
was fired on from below. The bullet passed
through the adductor muscles of the left thigh
and struck him on his shoulder with the effects he
has described. It has been objected that the ac-
count cannot be genuine because no marksman
would take up by choice so hazardous a position,
but this is surely to underrate the intrepidity and
agility of the hardy Afghan mountaineer" (103).

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

143

Another conjecture is that, in fact, "the wound
was located half way between" the shoulder and
the leg, actually in the groin, but Watson could
not mention it in deference to the sensibilities of
the reigning queen, Victoria (107). Will an expla-
nation be forthcoming based on Godel's theorem
of paradoxical undecidability?

Doyle wrote articles on medicine, but it is his
fiction that is replete with references to medicine.
In his non-Sherlock Holmes stories, over one hun-
dred doctors appear (6, 86, 108). References to
medicine in the Holmes stories have been tabu-
lated, with note of the real patients he had en-
countered together with their fictional counter-
parts (89, 90). In the 60 Holmes adventures, there
are 68 diseases, 32 medical terms, 38 physicians,
12 medical specialties, 22 drugs, six hospitals,
three medical journals, and two medical schools
(89, 90). They include descriptions of infectious
diseases, cancer, gout, trauma, pulmonary dis-
ease, cardiovascular disease, ocular disease, geri-
atrics, insanity, and neurologic disease. The med-
ical descriptions are sometimes vivid, like the
man afflicted with gout who "could chalk his bil-
liard cue with his knuckles" (89), presumably by
extruding the tophaceous deposits.

Holmes has been claimed by several medical
specialties, including rheumatology, dermatol-
ogy, forensic pathology, ophthalmology, neurol-
ogy (97, 109, 110); and it has been hinted that all
the while he practiced semiotics, interpreting
signs and symbols (111). Holmes is also regarded
as a cognoscente of heredity, an interest stimu-
lated in part by a close resemblance to his
brother, who is described as even more misan-
thropic than he, a member of a club in which "no
member is permitted to take the least notice of
any other one" (112).

One can argue that pharmacology was
Holmes' specialty, since 22 different drugs are
mentioned in the stories (90), as well as 36 con-
ditions due to drugs or toxins (97), including
Holmes' use of cocaine. Watson's plea to Holmes
to quit cocaine is pharmacologically sound:
"Count the cost! Your brain may, as you say, be
roused and excited, but it is a pathological and
morbid process, . . . and may at last leave a per-
manent weakness. You know, too, what a black
reaction comes upon you. Surely the game is
hardly worth the candle. Why should you, for a
mere passing pleasure, risk the loss of those great
powers with which you have been endowed?" (88).
Doyle himself was interested in drugs. He re-
ceived second-class honors in Materia Medica and
Therapeutics as a student (108). In a letter to the
British Medical Journal while a medical student,

Doyle described his self-experimentation with
gelsemium (113), a plant containing the stimu-
lant gelsemine. He wrote comments on the mar-
gins of his textbook of therapeutics, at least one of
which, on opium, is correct: "I'll tell you a most
serious fact / That opium dries a mucous tract /
And constipates and causes thirst / And stimu-
lates the heart at first" (88, 108). He also said,
"Wondrous was the science which combined so
many powerful drugs, and yet so accurately bal-
anced them that they never modified the action of
each other" (108, 114), a statement that would be
revised today. Yet in an address to medical stu-
dents in 1910, Doyle did not flatter therapeutics,
likening its practice to a blind man's swinging a
club at random, sometimes hitting the disease,
sometimes the patient, the pharmacopoeia being
the club (108, 114). However, almost all the drugs
mentioned in the Holmes stories hit the signs and
symptoms, though not the disease.

Since almost all the drugs affect the nervous
system, if Holmes must be assigned a medical
specialty, neurology can make the strongest
claim. Moreover, the neurologic conditions men-
tioned in the Holmes stories have been collected
(97, 109, 110): all four novels and 73% of the 56
short stories contain statements about a neuro-
logic condition, totaling 104 examples. These ob-
viously reflect the interest of Doyle, whose MD
thesis was on tabes dorsalis (87).

Doyle's Non-Holmesian Interests. Besides
Sherlock Holmes stories, Doyle wrote science fic-
tion, short stories, including those on medical top-
ics which have been collected (88, 108), historical
romances, adventure stories, tales of boxing, po-
ems, biography, newspaper reports about his vis-
its to all fronts in World War I, a comic opera
titled Jane Annie with James Barrie, and works
on spiritualism (6). Doyle, irritated by the con-
stant demands for more Sherlock Holmes stories,
ended Holmes' life in 1893, only to feel compelled
by readers to resuscitate him 10 years later.
Doyle regarded the Holmes stories as "a lower
stratum of literary achievement" (102). Doyle's
statement, "if I had never touched Holmes, . . .
my position in literature would at the present mo-
ment be a more commanding one" (89) may well
be correct. His historical adventure, Micah
Clarke, published two years after his first Holmes
story, was praised by Oscar Wilde and other lit-
erary figures (6). Doyle's medical stories placed
him "in the front rank of living writers," accord-
ing to a review in a New York newspaper, and
another American newspaper wrote, "No series of
short stories in modern literature can approach
them" (108).

144

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

Doyle's other accomplishments have been
noted (6, 88, 91, 92, 108). Holmes' concern with
mud, dust, stains, footprints, tobacco ash was new
to crime-solving. The first report of identification
of typewriting is due to Doyle. Before technical
reports appeared on the subject, he commented on
powder marks' defining the distance at which a
gun was discharged (6). Alphonse Bertillon, who
has been called the creator of forensic science, ad-
mitted his debt to Holmes and recommended his
methods to police (6). Doyle's book about the An-
glo-Boer War is regarded as a standard work (91).
After visiting the fronts in 1918, he wrote a six-
volume work on the British campaign in France.
He was largely responsible for the British Navy's
introduction of safety vests and collapsible floats
(91). He was an early advocate of the Channel
Tunnel. He founded a group to promote Anglo-
American relations. He was president of the Di-
vorce Law Reform Union. His pamphlets The
Crime of the Congo helped force the Belgian gov-
ernment to reform the treatment of the natives
(91), described by Doyle as "robbed, debouched,
mutilated and murdered" (88); the pamphlets
were published at his own expense, and profits
went to the University of Edinburgh for a schol-
arship fund (92). His short paper describing the
typhoid epidemic during the Boer War begins
with a tribute to three groups who put in "more
solid and unremitting toil than any others," and
they were those who provided the food, those who
worked on the railroad, and those who served as
medical orderlies (115). He was revolted by cru-
elty to animals, while he castigated the "antihu-
man campaign" of the antivivisectionists, point-
ing out the many lives saved by inoculations
developed in animals (88). He wrote letters to ed-
itors advocating vaccination against smallpox
(88). He was nearly correct in his prediction in
1883 that "It is probable that in the days of our
children's children, consumption, typhus, ty-
phoid, cholera, malaria, scarlatina, diphtheria,
measles, and a host of other diseases will have
ceased to exist" (89). In 1890, after visiting Berlin
to appraise the claim that Koch's tuberculin was
a cure for tuberculosis, he questioned, iconoclas-
tically and correctly, its efficacy (93, 108).

He wrote to Lancet and the British Medical
Journal in 1907 requesting opinions from oph-
thalmologists about the imprisonment of a se-
verely myopic man found guilty, on the basis of
anonymous letters, of mutilating animals (116):

Sir — Might I ask you in the cause of justice to permit me to
put the following question to those of your readers who are
engaged in eye practice? Do you consider it physically pos-
sible for Mr. George Edalji, whose degree of myopic astig-

matism as determined by retinoscopy under homatropine is
[the numbers are given] to have set forth without glasses on
a pitch dark night with neither moon nor stars; to have
crossed country for half a mile, climbing fences, finding
gaps in hedges, and passing over a broad railway line; tc
have found and mutilated a pony which was loose in a large
field, to have returned half a mile, and to have accom-
plished it all under thirty-five minutes, the limit of the
possible time at his disposal?

A consensus of scientific opinion upon this point would
greatly aid me in getting justice for this young professional
man, condemned for an offense which in my opinion he
could not possibly have committed.

The man was freed. He attended Doyle's wedding
(88, 108).

In 1910, Doyle told medical students wha1
the medical profession offers (114): "that you ar€
the friends of all, that all are better for your lives
that your ends are noble and humane. That uni
versal goodwill without, and that assurance oi
good work within, are advantages which cannol
be measured by any terms of money. You are the
heirs to a profession which has always had highei
ideals than the dollar. Those who have gone be
fore you have held its reputation high. Unselfish
ness, fearlessness, humanity, self-effacement
professional honour — these are the proud quali
ties which medicine has ever demanded from hei
sons. They have lived up to them. It is for yoi
youngsters to see that they shall not decline dur
ing the generation to come." A book was dedi
cated to Doyle "whose fundamental integrity
ubiquitous optimism, boundless vigor, inheren'
compassion, animated humor, prescient imagina
tion, and exceptional writing ability have giver
to future generations legacies in many genres
not the least of which is humanistic medicine pro
mulgated through fictional works" (108).

In his last piece of fiction, published in 1929
Doyle commented on the mores of the times
"There was no longer . . . the cultivation of th(
mind, but we had a glimpse of a people who wen
restless and shallow, rushing from one pursuit t(
another, grasping at every pleasure, for evei
missing it" (91). A year later he died, spared hav
ing to witness yet further degeneration.

The epitaph on his tombstone reads "Stee
True, Blade Straight" (92, 108).

Physician-Writers Who
Continued Medical Practice

The list of physician-writers who did not be
come medical truants is also long and imposing
Rufus Ephesius, physician to Cleopatra, wrote po
ems (11). The great Roman poet Virgil (70 BC-IJ
BC) was a physician (11). Girolamo Fracaston
(1483-1553) was a renowned physician, geologist

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

145

astronomer, and poet, who named and described
syphilis in a poem. Syphilis or the French Disease
(12, 67), summarizing all that was known about
the disease. Ambroise Pare (1510-1590) was sur-
geon to four kings and wrote sonnets (11). Thom-
as Lodge (1558-1625) took his medical degree
when he was over 40, having been occupied as a
lawyer, soldier, sailor and writer of plays, novels,
pamphlets and poems (18). One of his works, Ro-
salynde, provided Shakespeare with the plot of As
You Like It. He wrote A Treatise on the Plague in
1603 while the disease was raging in London, and
died of the disease 22 years later (18). Thomas
Campion (1567-1620) was physician, poet, com-
poser and lutenist (11). Thomas Browne (1605-
1681) made "the harmony between the physician
and man of letters complete," revealing in "glow-
ing and original expressions, the poetic sap which
flows through all the minds of the age" (11). His
Religio Medici "was written to reveal the soul in
a physician" (11). Francesco Redi (1626-1698)
disproved spontaneous generation and wrote son-
nets (11). Samuel Garth (1661-1719) was consid-
ered the best poet of his time while engaging in
political activities and serving as a physician to
King George I, the Duke of Marlborough and
other notables. John Arbuthnot (1667-1735)
wrote the text for Handel's oratorio Ester while
serving as Queen Anne's physician (6). He intro-
duced John Bull in his writings. Arbuthnot was a
Scotsman, as was Tobias Smollett (1721-1771),
another physician-writer, whose novels influ-
enced Dickens (6). Albrecht von Haller (1708-
1777), whose ideas on tissue irritability provoked
great activity by Galvani and others, was a poet
and novelist too (11). Erasmas Darwin (1731-
1802), the grandfather of Charles Darwin, prac-
ticed medicine and wrote a long poem on Lin-
naeus' botanical classification (6). Rene Laennec
(1781-1826) invented the stethoscope and wrote
poetry (11).

Some Irish Nontruants. Ireland contributed
to the roster of nontruants as well as the roster of
truants (6, 7, 35). William Wilde (1815-1876) be-
came a luminary of European medicine before the
age of 40, honored by many universities. He
founded the Dublin eye hospital, wrote a standard
textbook on ear surgery and books on history, ar-
cheology, travel, fishing and other subjects. His
wife was a poet and wrote articles ardently sup-
porting Irish independence. After Wilde ended an
affair, the ex-mistress accused him of seducing
her under chloroform. Wilde's wife wrote a vitri-
olic letter to the woman's father, which engen-
dered a libel suit. Although the award to the lit-
igant was only titular because of her inconsistent

testimony, Wilde never recovered from the scan-
dal. His son Oscar's ending mimicked his (7).

Oliver St. John Gogarty (1878-1957) was
graduated with honors from Trinity College in
Dublin and also studied medicine in Oxford and
Leiden. He practiced surgery all his life while be-
coming, according to W. B. Yeats, "one of the
great lyric poets of our age" (6, 7, 35). He wrote
prose and limericks too. Gogarty was one of the
founders of the Sinn Fein, a senator of the Irish
Free State, "the wildest wit in Ireland" (6), a mas-
ter of repartee that was often acerbic, an aviator,
a bicycle racer, and a champion swimmer. His
athleticism was helpful. In 1923, he was kid-
napped by armed men and escaped by diving into
the river Liffey and swimming away under a hail
of bullets. Earlier he received a medal for diving
into the river to save a drowning man. By quick
thinking, he saved the life of Michael Collins, one
of the founders of the Irish Free State (6). He is
now remembered mostly as James Joyce's drink-
ing companion and the man portrayed as Buck
Mulligan in Joyce's Ulysses. His comments on
Joyce varied, sometimes disparaging (22), some-
times praising (35). He left Dublin for London,
died in New York City, and was buried in Conne-
mara. He was told he would not likely be praised
as a poet until 50 years after his death (6). In the
summer of 1992, Trinity College had an exhibi-
tion extolling its greats, and Gogarty received
scant mention (117). We wait, comforted by the
opinion that "there is no fear that his name will
be writ in water" (18).

Oliver Wendell Holmes and William Carlos
Williams. Two great American physician-writers
practiced both occupations all their lives. Oliver
Wendell Holmes (1809-1894), self-characterized
as a member of the "Brahmin caste of New En-
gland" and admired by Arthur Conan Doyle,
never left medicine and meantime wrote poems —
many related to medicine — three novels, and nu-
merous essays (118). William Osier considered
him "the most successful combination which the
world has ever seen of the physician and man of
letters" (118). While briefly a law student, he re-
lieved his boredom by writing poems, one of
which. Old Ironsides, resulted in the preservation
of The Constitution, a gallant ship that had been
marked for destruction. He went to Paris to study
medicine with Pierre Charles Alexandre Louis,
who propounded the practice of scientific medi-
cine. Holmes, adhering to Louis' principles, de-
cided "not to take authority when I can have
facts; not to guess when I can know." He received
the M.D. degree from Harvard in 1836. A major
contribution, in 1843, was The Contagiousness of

146

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

Puerperal Fever, a disease contracted during
childbirth because, he charged, physicians and
other participants failed to respect cleanliness.
He recommended procedures to avoid infection.
For making identical proposals, I. P. Semmelweis
suffered ridicule, abuse, and banishment from Vi-
enna and was driven to screaming on the street
"Wash your hands!" before proving his hypothesis
by dipping his cut fingers into an infected cadaver
in Budapest, fatally infecting himself, and dying
two months later in a mental hospital (65). On
puerperal fever, as on all subjects. Holmes' rhe-
torical gifts were trenchant: "the time has come
when the existence of private pestilence in the
sphere of a single physician should be looked
upon, not as a misfortune, but a crime," and "The
woman about to become a mother . . . should be
the object of trembling care and sympathy. . . .
God forbid that any member of the profession to
which she trusts her life . . . should hazard it neg-
ligently, unadvisedly, or selfishly." He became
professor of anatomy and physiology at Harvard
Medical School in 1847 and served for 35 years.
He persisted in demanding a mindful practice of
medicine, satirizing physicians who continued to
practice archaic procedures like blood-letting in a
poem titled Rip Van Winkle, MD., published in
1870. He was an effective and popular teacher,
holding to a motto of "iteration and reiteration."

The greatest American poet among physi-
cians, most would agree, is William Carlos
Williams (119, 120), who during a long life (1883-
1963) wrote essays, memoirs, stories, and the
most original and influential American poetry of
this century. For over 40 years he practiced in
Rutherford, New Jersey.

Since about 1875 the list of physician-writers
practicing both crafts is international. The
Canadian W. H. Drummond (1854-1907) was
dubbed "The Poet Laureate of British America"
(18). Another Canadian, John McCrae (1872-
1918), was killed in World War I (18), leaving us
the poem In Flanders Field. The Czech Miroslaw
Holub continued as a pathologist and writer
(121). Tomio Tada, a professor of immunology at
Tokyo University Medical School, recently wrote
and had produced a Noh play based on a heart
transplantation (122). Many physicians who
achieved fame in the biomedical sciences wrote
poetry and novels on the side; they include Ed-
ward Jenner, Charles Sherrington, Russell Brain,
Henry Head, S. Weir Mitchell, and Wilder Pen-
field (18, 22).

Heightened Current Activity. In the past two
decades, the number of active practicing physi-
cian-writers has increased remarkably. Much,

though not all, of the writing stems directly from
experiences in medicine. Among these physician-
writers are Danny Abse, Howard Brody,
Jonathan Cole, Bruce Dobkin, Colin Douglas,
Henry Eisenberg, Robert Goldwyn, David Hil-
fiker, Perri Klass, Harold Klawans, Arthur
Kleinman, Robert Marion, Russell Martin,
Fitzhugh Mullan, Michael O'Reilly, Oliver Sacks,
Kenneth Sanders, Leon Schwartzenberg, Richard
Selzer, John Stone, Lewis Thomas, Gerald Weiss-
man, and Irving Yalom.

Professional Groups and Publications. The
heightened activity (27, 42) is manifest in the for-
mation of the American Physician Poetry Associ-
ation and The Physician Author Society, which
publish a quarterly. The Physician Author, and
offer workshops and seminars that provide credit
for continuing medical education. Literature and
Medicine is a scholarly journal published by the
Johns Hopkins University Press, and a new one
has been announced, Mediphors, described as a
literary journal for the health professions. Physi-
cians' penchant for writing poetry is evident in
poems appearing weekly in the Journal of the
American Medical Association and frequently in
the New England Journal of Medicine, Perspec-
tives in Biology and Medicine, and The Pharos,
the journal of the medical honor society. The
William Carlos Williams Poetry Contest for Med-
ical Students is sponsored by the Northeastern
Ohio Universities College of Medicine, and some
of the poems submitted to this competition are
published in the Journal of the American Medical
Association.

Proposed Connections between
Medicine and Literature

The plethora of physician-writers raises the
question of the relation between literature and
medicine or other sciences. The dual activities do
not imply that medicine and literature have firm
bonds, although many would hope so. William
Wordsworth expected the sciences to become
"proper objects of the Poet's art," also declaring
"it is the passioned expression which is the coun-
tenance of all science" (123). Ezra Pound, perhaps
influenced by his friend William Carlos Williams
(119), is quoted as recommending that the proper
method to study poetry is the method of contem-
porary biologists. John Dryden believed that a
complete poet should be schooled in science (124),
Matthew Arnold that science would be incom-
plete without poetry (19), and Robert Glynn that
"a physician should be a kind of poet" (18). Sam-
uel Coleridge hoped for communication between

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

147

men in science and literature (124), anticipating
a union of science and poetry (123).

It has been stated (125) that science and po-
etry, the "two great ways of seeing lie on the same
imaginative continuum. They do not compete;
they connect." A highly literate, highly gifted sci-
entist (126) maintains that they do compete, they
cannot be reconciled, and when one arrives, it ex-
pels the other. He believes that though "imagina-
tion is the energizing force of science as well as
poetry" and though "they start in parallel . . .
they diverge," because "in science imagination
and a critical evaluation of its products are inte-
grally combined," acting in synergy. Both poetry
and science begin as stories, he asserts, but the
stories differ in their purposes and in the kinds of
evaluations of them. "All scientific theories must
make sense . . . but in addition they are expected
to conform to reality, to be empirically true." In
this essay (126), science and literature are differ-
entiated, and neither is derogated.

There is no yield in arguing whether or how
closely literature and medicine are connected.
Neither is harmed standing independently, and
both are accessible. Chekhov, in answer to a let-
ter (83) in which science was denigrated, de-
manded respect for both (May 15, 1889):

You speak of the right of these or those kinds of knowledge
to exist, whereas I speak not of a right but of peace. I want
people not to see war where there is none. The different
kinds of knowledge have always abided in peace with one
another. Anatomy and belles lettres have an equally illus-
trious origin, the very same goals, the very same enemy —
the devil — and they have absolutely nothing to wage war
about. There is no struggle for existence among them. If a
man knows the theory of the circulation of the blood he is
rich; if, over and beyond that, he masters the history of
religion and the lyric ... he becomes not poorer but richer,
ergo, what we are dealing with is only pluses. That is why
geniuses have never waged war, and why, in the case of
Goethe, the naturalist managed to live serenely side by side
with the poet.

The view attributed to the genius playwright
near-physician Biichner is that science and art
simply represent "different perspectives from
which to view the same subject" and "different
languages with which to describe it" (16).

There is no question that the medical experi-
ence has influenced the literature of some physi-
cians. Many have explicitly used medicine in
their literature. In his obituary for Schnitzler,
Thomas Mann wrote that Schnitzler's literary
sensibility was "sharpened by the experience of a
doctor" (84). Although assiduous searching is
needed to find reference to medicine in Keats' po-
ems, it has been concluded that "medical knowl-
edge was . . . one of the fountains that flowed into

. . . his poetic mind, bringing with it ideas and
images" (80). Others praised the scientific method
that medicine taught them. Chekhov wrote in a
letter of November 1888 (83): "Those who have
mastered the wisdom of the scientific method and
are able to think scientifically experience many
charming temptations. . . . Those who possess the
scientific method feel with their souls that a mu-
sical composition and a tree have something in
common, that both are created in accordance with
equally regular and simple laws." On October 11,
1889, he wrote (83):

I don't doubt that the study of the medical sciences seri-
ously affected my literary work; they significantly enlarged
the field of my observations, enriched me with knowledge,
the true value of which for me as a writer can be understood
only by one who is himself a physician; they also had a
directive influence and probably because I was close to med-
icine I avoided many mistakes. Acquaintance with the nat-
ural sciences, with the scientific method, kept me always on
guard and I tried, wherever possible, to bring my writings
into harmony with scientific data, and where this was im-
possible, I preferred not to write at all. Let me observe that
creativity in the arts does not always permit total agree-
ment with scientific data; thus it is impossible to represent
on the stage death from poisoning as it actually takes place.
But agreement with scientific data must be felt in the con-
ventions accepted, that is, it is necessary for the reader or
spectator to grasp clearly that these are only conventions,
and that he is dealing with an author who knows the true
facts. I do not belong to the fiction writers who have a
negative attitude toward science, nor am I one of those
artists who think that they can arrive at everjrthing by the
intellect alone. I would not want to be one of them.

Arthur Conan Doyle gave similar testimony, tell-
ing medical students that medical education
"tinges the whole philosophy of life and furnishes
the whole basis of thought. The healthy skepti-
cism which medical training induces, the desire
to prove every fact, and only to reason from
proved facts — these are the finest foundations for
all thought" (89, 108, 114).

"Romantic Science" and the "Grim Ro-
mance of Medicine." If education in the scientific
method were a major determinant of the associa-
tion between writing fiction and an education in
medicine, then a roster of novelists and poets
would be rich in physicists and philosophers. The
physician-writer association must have addi-
tional or other reasons. Pervading many (though
not all) of the writings of physicians, as Oliver
Sacks has noted (127), is what A. R. Luria de-
scribed as romantic science, where the author pre-
sents "a patient ... in his wholeness, while delin-
eating simultaneously the intimate structure of
his being," an "analytical description with a
deeply personal empathetic feeling for the sub-
jects" where "the syndrome ... is embedded in a

148

THE MOUNT SINAI JOURNAL OF MEDICINE

March 199;

person. . . . And these are conjoined — the syn-
drome is always related to the person and the
person to the syndrome — the personal and the sci-
entific are always, hopefully, fused" (127).

In Doyle's speech to medical students in 1910
(89, 108, 114), he described medicine as a "grim
romance," using romance to mean narratives that
are marked by the emotional and imaginative ap-
peal of human experience. It insufflates his writ-
ings and shows what is increasingly called "hu-
manistic medicine." In his M.D. thesis on tabes
dorsalis, Doyle wrote, "As the disease progresses
the sufferer gets some relief from pain, the sud-
den shocks dying away and being replaced by an-
algesia. . . . Slowly the unfortunate victim sinks
from one gradation of misery to another and can
only look forward to the death which may reach
him from pure exhaustion or may come from the
involvement of the vital centres in the medulla."
In recounting the influence of disease on history,
he told the students that Napoleon's

six years at St. Helena furnish a clinical study of gastric
disease which was all explained in the historical post mor-
tem, which disclosed cancer covering the whole wall of the
stomach, and actually perforating it at the hepatic border.
Napoleon's whole career was profoundly modified by his
complaint. There have been many criticisms ... of his
petty, querulous and undignified attitude during his cap-
tivity; but if his critics knew what it was to digest their food
with an organ which had hardly a square inch of healthy
tissue upon it, they would perhaps take a more generous
view of the conduct of Napoleon. ... I think that his forti-
tude was never more shown than during those years — the
best proof of which was, that his guardians had no notion
how ill he was until within a few days of his actual death"
(108).

In reply to a letter questioning the need for
stark realism in his medical stories, Doyle wrote,
"If you deal with this life at all ... it is quite
essential that you should paint the darker side,
since it is that which is principally presented to
the Surgeon or the Physician. He sees many beau-
tiful things, it is true; fortitude and heroism, self-
sacrifice and love, but they are all called forth (as
our nobler qualities are always called forth) by
bitter sorrow and trial. One cannot write of med-
ical life and be merry over it" (108). Doyle pre-
sented one of the doctors' dilemmas experienced
in his practice: "There was a grocer that devel-
oped epileptic fits, which meant butter and tea for
us. Poor fellow, he could never have realized the
mixed feelings with which I received the news of
a fresh outbreak. ... It was a ghoulish compact,
by which a fit to him meant butter and bacon to
me, while a spell of health for him sent me back to
dry bread" (108).

Maugham too recorded the emotive effects of
his experiences as a medical student (6):

For here I was in contact with what I most wanted, life ii
the raw. In those three years I must have witnessed prett;
well every emotion of which man is capable. It appealed ti
my dramatic instinct. It excited the novelist in me. Evei
now that forty years have passed I can remember certaii
people so exactly that I could draw a picture of them
Phrases that I heard then still linger on my ears. I saw ho\
men died. I saw how they bore pain. I saw what hope lookei
like, fear and relief; I saw the dark lines that despair drev
on a face; I saw courage and steadfastness. I saw faith shin
in the eyes of those who trusted in what I could only thinl
was an illusion, and I saw the gallantry that made a mai
greet the prognosis of death with an ironic joke because h
was too proud to let those about him see the terror of hi
soul. ... I do not know a better training for a writer than t
spend some years in the medical profession." He added in
letter that same year: "I can only wish that I had remaine
a doctor for three or four years instead of writing book
which have long been dead as mutton (6).

William Carlos Williams had similar feeling
after nearly 50 years of practicing medicine whil
writing, mostly poetry (120): "as a writer I havi
never felt that medicine interfered with me bu
rather that it was my very food and drink, thi
very thing which made it possible for me to write
Was I not interested in man? There the thing was
right in front of me. I could touch it, smell it. I
was myself, naked, just as it was, without a lie
itself to me in its own terms." And "when th
inarticulate patient struggles to lay himself bar
for you ... he reveals some secret twist of a whol
community's pathetic way of thought ... a man i
suddenly seized again with a desire to speak o
the underground stream which for a moment ha
come up just under the surface . . . there is n
better way to get an intimation of what is goini
on in the world." And "the hunted news I get fror
some obscure patient's eye is not trivial. It is pro
found: whole academies of learning, whole eccle
siastical hierarchies are founded upon it."

Doyle extolled

the value of kindliness and humanity as well as of know'
edge. ... A strong and kindly personality is as valuable a
asset as actual learning in a medical man. I do not mea
that trained urbanity which has been called "the bedsid
manner," but the real natural benevolence which . . . car
not be simulated by any forced geniality. I have known me
in the profession who were stuffed with accurate knowledg
and yet were so cold in their bearing, and so unsympatheti
in their attitude . . . that they left their half-frozen patient
all the worse for their contact. While, on the other hand,
have known . . . men who were of such exuberance of vita
ity and kindliness that . . . the mere clasp of their hand an
light in their eyes have left their patient in a more cheer
and hopeful mood (89, 108, 114).

These words are responses to suffering cou
pled with ideals. They voice the response to th^
patient not as a disease but as an ill person wit)
a disease entwined with a personal, emotiona
life. Even if all — Luria, Doyle, Maugham

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

149

Williams, and the "secular saint" Chekhov — were
explicitly taught this attitude in medical school,
it is almost certain that not all their confreres
became vested with the same feelings. All medi-
cal students at Edinburgh were urged by Joseph
Bell, "Be kind and gentle" (99). Doyle, according
to a classmate, was impressive "by his very kind
and considerate manner towards the poor people
who came to the out-patient department, whom
I'm afraid some of us were in the habit of treating
somewhat cavalierly" (90). In contrast, Doyle's
schoolmate, George Budd, advised, "Break your
patients in early, and keep them well to heel.
Never make the fatal mistake of being polite to
them" (91).

Very likely, each brought their feelings with
them to medical school. Maybe the sensibilities
of some writers allows them to experience em-
pathic reactions as physicians. For "empathy
has an esthetic component" (128). All were writ-
ers first; each was blessed with imagination.
Doyle and Chekhov published while medical stu-
dents. Maugham said that his only interest before
medical school was art and literature (43).
Williams wrote, "to treat the patient as a work of
art made him somehow come alive to me. . . . The
poem springs from the half spoken words of . . .
patients" (119). Williams' son wrote (119) that his
father sought a medical education for the income
it would bring to pursue art, that medicine was
secondary to poetry in his favor. Even while busy
with medical practice, Williams wrote his mother
that he had given up medicine for poetry; this
statement — rhetorical, since he continued to
practice — did not diminish his professionalism or
his capacity for empathy. He wrote (119) of
"adopting the patient's condition as one's own to
be struggled with toward a solution. ... I actually
became them ... so that when I detached myself
from them ... it was as though I were reawaken-
ing from a sleep." The empathetic feeling that
medical schools would aspire to inculcate were
seemingly intrinsic to their persons. Heroes can-
not be imitated. But the sick would benefit if peo-
ple with similar compassion and empathy could
be selected for medical schools — if they could be
confidently identified, and they cannot.

Pleas for a more humane, compassionate, em-
pathic medicine have been made for centuries, in-
cluding pleas by such great clinicians and teach-
ers as Oliver Wendell Holmes, Osier, and
Peabody (129). "To their patients what are the
most scientific physicians if they know all things
save the human heart?" (19). The ceaseless pleas
clearly imply that the lack and the need are not
new. In the past two decades, the entreaties ap-

pear like paroxysms of revelation and have grown
to adjurements, presented by physicians and oth-
ers (128, 130-133), often emotionally. We are rue-
fully asked if patients are seen as "human loaves
of bread in the health-care supermarket?" (134).
We are told, "We have priced compassion out of
the American soul" (135). A hospital floor de-
signed to treat a "patient like a person" and to
"determine whether there is benefit to being nice
to people" provokes the mock epiphany, "Caring.
What a concept!" (136).

The increasingly vociferous outcry by physi-
cians could suggest that medical practice has be-
come less humanistic, rather than that patients
have become more demanding. Maybe simple ci-
vility is under erasure, but other reasons have
been proposed. Noncompassionate medicine has
been attributed to increasing regulatory pres-
sures that cause distractions from patient care, to
a "craze for biomedical tests," and "to a reim-
bursement system that rewards costly technical
procedures rather than clinical assessment"

(133) . For technical procedures, "we can bill the
government . . . because they are easier to quan-
tify than judgment, compassion and kindness"

(134) . A psychiatrist, deprived of reimbursable
technology, has facetiously described experi-
ments on a laser-based "psychoscope" that en-
ables the ego, superego, and id to be separately
visualized (137). In like fashion, another de-
scribes a PET scanner to examine fluctuations in
Descartes' seat of the soul, the pineal body, a
scanner that

can be guaranteed to increase the earning capacity of
American psychiatrists. . . . Because its immediate imag-
ing properties make it ideal for office use, psychiatrists
rather than radiologists can bill for it, and thus attain eco-
nomic parity with other technology-based disciplines. Be-
cause it will no longer be necessary to inquire of the patient
about the state of his or her soul, a notoriously time-con-
suming, uncertain and altogether messy procedure, time
savings will be prodigious; productivity will increase; its
cost-effectiveness will please both supply side and free mar-
ket economists (132).

(Testing also provides data for defense against li-
tigious assaults, as well as for diagnosis.)

Other reasons that have been offered for the
apparent dwindling of compassion are that house
officers and medical students are overworked and
that too much science is taught in medical school.
However, house officers have always been over-
worked, and they now have shorter hours than
they had fifty years ago. Medical students have
been overworked throughout modern times. Re-
calling his education over a hundred years ago,
Doyle told students (89, 108, 114) that a medical
curriculum

150

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

sets a very high standard of strenuous work ... it remains
a precious heritage for life. To the man who has mastered
Grey's [sic\ Anatomy, life holds no further terrors. To re-
member that huge catalogue of attachments and ganglia
and anastomoses is, I think, the most arduous task any of
you will ever be called upon to do. But it is worth doing, for
it leaves you with that ideal. In after life, when you are
confronted with any task, you may say to yourself, "Well, I
will do it as thoroughly and accurately as I did my anatomy
at College" and if you do it in that spirit . . . nothing and
nobody can stand against you. All work seems easy after
the work of a medical education.

Attributing the ebbing of compassion to too
much science in the medical curriculum is
equally unconvincing. The bulk amount of sci-
ence taught today in American medical schools is
no greater than it was fifty years ago. It is the
variety of science that has increased. Burgeoning
sciences such as molecular biology are now appro-
priately included in the curriculum, but the time
spent on these has been taken from other sci-
ences; less time is given to gross anatomy, and
students are not required to learn the anatomic
details previously demanded. Moreover, at most
American medical schools, the total scheduled
teaching hours have fallen greatly over the past
fifty years. One can argue that the students are,
as always, burdened with too many facts while
too little time is spent in teaching critical think-
ing— the ability to analyze data and to probe the
validity of an inference. These and other changes
in curriculum may well be indicated, but the tech-
nological advances in diagnosis and treatment
rest on science. The teaching of science may need
modification, but its abasement runs the risk of
fostering ignorance, arresting the benefits that
science brings, and, therefore, menacing patient
care. It has been convincingly argued (133) that
reducing the biological sciences will not teach
physicians "to attend to the patient as well as to
the disease." The objective is to "reconcile scien-
tific understanding with human understanding"
(130). What may be useful is additional science
that relates to patient care — instruction in the
social sciences (133) and effective instruction in
the behavioral sciences (138).

Another suggestion is that recruiting stu-
dents who majored in humanities to medical
school will enhance humane medicine. Few would
argue with Abraham Flexner's early proposal
(139) that physicians should be educated in the
humanities because descriptive skills and lan-
guage could help to individualize patients. Hu-
manities majors can add heterogeneity to a class,
itself educative, and could make the class more
interesting to teach. However, to presume that
humanities majors are more humane is to trans-

duce incoherently words of the same origin,
rather like presuming that sociology majors are
more social. Anyone working at a university
would find difficulty in discerning a difference in
the humanism displayed by a professor in the hu-
manities and a professor in the sciences.

The humanities, especially literature, have
been used in schools of medicine in the past
twenty years, and more recently in law as well, in
law to produce a new word, poethics (140). The
purpose is not only to teach ethics, but, concomi-
tantly, to increase the practice of humane medi-
cine (38, 141-143). It has been emphasized that to
be effective, the humanities should be taught in
the context of medicine, integrated with medicine
(144).

The "mere force of the imagination," David
Hume wrote (60), could make a malady real and
help students to experience, or at least to under-
stand, the patient's feelings (130, 131, 145). The
poet Percy Bysshe Shelley said, "A man, to be
greatly good, must imagine intensely and com-
prehensively; he must put himself in the place of
another and of many others; the pains and plea-
sures of his species must become his own. The
great instrument of moral good is the imagina-
tion" (47). Even if Williams' habit of "adopting
the patient's condition as one's own" were advis-
able, not everyone can do it. It is enough to un-
derstand the patient's feelings, to gain access to
the feelings, maybe by recalling one's own feel-
ings in an analogous situation or by imagining
what one's feelings would be (145), a practice used
by actors to understand a character in a script.

Carried further is the assertion that for
imagination to provoke empathy, it must be
linked to memory: "He who has never suffered is
without empathy; it is as simple as that" (47).
Samuel Johnson, in his biography of the great
Dutch physician Hermann Boerhaave, asserted
that Boerhaave's "own pain taught him to com-
passionate others" (33). An extreme of this view
was offered by the French essayist Montaigne,
who is quoted (146) as saying, "So Plato was right
in saying that to become a true doctor, the candi-
date must have passed through all the illnesses
that he wants to cure and all the accidents and
circumstances that he is to diagnose. It is reason-
able that he should catch the pox if he wants to
know how to treat it." The goal of suffering may
be ethically and readily achieved by admitting
medical students in disguise to a hospital, a prac-
tice that has recently been instituted for house
staff to increase sensitivity (147).

But literature as a teaching tool may be pref-
erable to prescribed suffering. Literature, like the

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

151

other arts, can evoke vicarious experiences by en-
larging the imagination, show the uniqueness of
people, and reveal importances by its ability to
fabricate orderliness out of the obscuring random-
ness and chaos of living (29, 31, 141, 142, 148-
150). Literature courses and reading lists are of-
fered at some medical schools (30, 31, 141, 142,
151-153). In some courses, a literary expert, an
ethicist, and a philosopher help the clinician to
conduct the sessions (153). The readings are
firmly related to clinical experiences (141, 152,
153). Selections are also made to show how sex,
social class, cultural background, education, reli-
gion and race affect responses of patients and cli-
nicians. A literature exists by women describing
their health problems (154, 155). Literature that
is devoid of political rhetoric is available about
the mores, values, attitudes, and feelings of
working-class people, as in James T. Farrell's
Studs Lonigan; of blacks, as in Richard Wright's
Native Son, the novels of Toni Morrison, and a
recent collection of prose and poetry (156); of im-
migrants, as in Henry Roth's Call It Sleep. All
could be enlightening to medical students, most of
whom are still largely from the privileged class.
They might learn that to some working-class pa-
tients, having a disease is to be a loser; that com-
plaining about pain is not acceptable; and that
some medical expressions are foreign to them:
there is no diabetes in the family, but there is
"sugar"; and he never had gonorrhea, but he did
have the clap.

The visual arts (60, 161-164) — painting, car-
icatures— could also be useful in enhancing un-
derstanding and empathy. Holbein's painting The
Body of the Dead Christ in the Tomb overwhelmed
Dostoevsky. His wife recalls, "He stood for twenty
minutes before the picture without moving. On
his agitated face there was the frightened expres-
sion I often noticed . . . during the first moments
of his epileptic fits. He had no fit . . . but he could
never forget the sensation he had experienced
. . . : the figure of Christ . . . whose body already
showed signs of decomposition, haunted him"
(165). At the least, the visual arts may be useful
in enhancing the capacity to observe.

The Patient's History. The immediate rela-
tionship between medicine and literature, repeat-
edly pointed out, is that a patient's history and
the progress of the illness constitute a narrative.
It has even been subjected to the techniques of
literary analysis (44). The initial history is "an
inquiry," which, a historian writes, "is surely to
require the telling of stories" (157). The stories
require attentive listening; nearly half the pa-
tients in one survey believed that their physi-

cians do not listen (158), maybe because most of
them were raised on television rather than radio.
Nevertheless, repeated studies have shown that
the history alone, without the help of physical
examination or laboratory studies, leads to
the final diagnosis in more than half the pa-
tients (159) — not necessarily the half that are lis-
tened to.

Despite this success, some have decried that
history-taking today has drifted from concern for
the patient's thoughts and feelings. It has been
suggested that physicians (and students) must
stimulate the patient's narrative flow (145),
which may, as Anatole Broyard wrote (160), help
the patient enter the physician's heart. What may
be especially revealing is a deeper anamnesis, to
have a patient describe how he or she feels about
a disease (130, 131) — how does a patient with ce-
rebral palsy, or one who has been different in
other ways since childhood, feel about having the
infirmity and how it influenced his or her life? — a
testimony providing a lesson in somatopsychia-
try.

Oliver Sacks

Oliver Sacks is an accomplished practitioner
of Luria's romantic science, of Doyle's romantic
medicine. Oliver Sacks was described by Pauline
Kael (166) in her review of "Awakenings," the
movie based on his book:

The neurologist Oliver Sacks is a supremely odd and com-
plicated man. This bearded giant, a motorcyclist, a fern
lover, and a weight lifter (who has been known to squat-
press six hundred pounds), has a passionate curiosity about
people. Born in London, in 1933, of physician parents who
trained in neurology, and with two older brothers who be-
came physicians, he grew up in a household where every-
body swapped medical stories, and he has developed the
clinical case study into an art form. Published as collections
of essays, his narratives about his patients are speculative,
exploratory, and maybe a little mad. He's drawn into his
patients' aberrant states — he sees their abnormalities as
brilliantly unique forms of consciousness to be charted —
and, as he attempts to interpret them, he's drawn in further
and further. He never closes off a subject; he goes on raising
more possibilities, and his emotional essays sprout foot-
notes— excursions into ideas he's trying out. Reading him,
you're absorbed in the play of his mind. (You may suspect
that he is, too — that he's half bewildered by himself.) He
probably makes more outre reaches of the imagination
than any other essayist, and you might think you were
reading freak-show detective stories if it weren't for the
transforming power of his childlike sincerity and his tough-
mindedness. Following him down a byway in a footnote,
you think. This is loony, and then, as you read more of the
tiny print, it doesn't seem so farfetched — it seems marvel-
ously plausible. You entertain the idea, conscious that his
interests are encyclopedic and he could go on footnoting
forever. . . .

A few months before the literary critic Anatole Broy-

152

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

ard died last year, he wrote, "My ideal doctor . . . would
resemble Oliver Sacks. ... I can imagine Dr. Sacks entering
my condition, looking around at it from the inside like a
benevolent landlord with a tenant, trying to see how he
could make the premises more livable for me. He would see
the genius of my illness. He would mingle his daemon with
mine; we would wrestle with my fate together."

Maybe the outstanding convergences of med-
icine and literature that Holmes, Doyle,
Maugham, Luria, Williams, and others displayed
are concisely put by Kael's "passionate curiosity"
to be "drawn into . . . patients' aberrant states."
Oliver Sacks is one who lives the convergences.

Acknowledgments

The author thanks Lance Garmer for translating the informa-
tion about Friedrich Wolf and for providing books of Gottfried
Benn; and Lina Mazzella for help with the manuscript.

References

1. Moynihan Lord. Truants: the story of some who deserted

medicine yet triumphed. London: Keynes Press, 1936.

2. Monro TK. The physician as man of letters, science and

action. London: Livingstone, 1951.

3. Cooke AM. Out of school. J Roy Coll Physns Lond 1974;

9:51-62.

4. Grifflts J. Call me a doctor: tales of six medical men who

went elsewhere and did other things. Arlington:
American Spectator, 1991.

5. Lieberson G. Medicine, mind and music. The CBS Leg-

acy Collection. CBS Records, 1971.

6. Smithers DW. This idle trade: on doctors who were writ-

ers. Tunbridge Wells: Dragonfly, 1989.

7. Sheridan N. Doctors and literature. Br Med J 1978; 299:

1779-1780.

8. Gifford Jr GE. Physician signers of the Declaration of

Independence. New York: Neale Watson Academic,
1976.

9. Jeffreys MVC. John Locke. Br Med J 1974; 4:34-35.

10. Osier W. An address on John Locke as a physician. Lan-

cet 1900; 2:1116-1123.

11. Webb GB. The role of the physician in literature. Medi-

cal Life 1929; 192-218.

12. Rodin AE, Key JD. Medicine, literature and eponyms: an

encyclopedia of medical eponyms derived from literary
characters. Malabar: Robert E. Krieger, 1989.

13. Spiegl F. The medical muse. Br Med J 1990; 301:1454-

1456.

14. Lock S. Roy Calne. JAMA 1991; 266:1736.

15. Eloesser L. Pirate and buccaneer doctors. Ann Med Hist

1926; 8:31-60.

16. Richards DG. George Biichner and the birth of modern

drama. Albany: State University of New York, 1977.

17. Dana CI. Poetry and doctors. Woodstock: Elm Tree, 1916.

18. RoUeston H. Poetry and physic. Ann Med Hist 1926; 8:

1-19.

19. RoUeston H. Some worthies of the Cambridge Medical

School. Ann Med Hist 1926; 8:331-346.

20. McDonough ML. Poet physicians: an anthology of med-

ical poetry written by physicians. Springfield: Charles
C Thomas, 1945.

21. Anonymous. Medical poets. Br Med J 1945; 2:698.

22. Ready WB. Medicine and literature: doctors in both fac-

ulties. Bull Med Lib Assoc 1962; 50:57-66.

23. Gordan JD. Doctors as men of letters: English and Amer-

ican writers of medical background — an exhibition in
the Berg collection. Bull NY Public Library November
1964.

24. Sergeant J, Sergeant H. Poems from hospital. London:

Allen and Unwin, 1968.

25. Sergeant H. Poems from the medical world. Lancaster:

MTP Press, 1980.

26. Peyre H. The doctor as man of letters. In: Peschel ER, ed.

Medicine and literature. New York: Neale Watson Ac-
ademic, 1980; 39-47.

27. Borgenicht L. American medicine: an annotated bibliog-

raphy. N Engl J Med 1981; 304:1112-1117.

28. Rousseau GS. Literature and medicine: the state of the

field. ISIS 1981; 72:406-424.

29. Trautmann J. Healing arts in dialogue: medicine and

literature. Carbondale: Southern Illinois Univ, 1981.

30. Trautmann J, Pollard C. Literature and medicine: an

annotated bibliography. Pittsburgh: University of
Pittsburgh, revised edition 1982.

31. Moore AR. The missing medical text: humane patient

care. Melbourne: Melbourne University, 1978.

32. Smithers D. On some medicoliterary alliances. Br Med J

1985; 291:1796-1801.

33. Rousseau GS. Literature and medicine: towards a simul-

taneity of theory and practice. Lit Med 1986; 5:152-
181.

34. Mukand J, ed. Sutured words: contemporary poetry

about medicine. Brookline: Aviva, 1987.

35. Lyons JB. Thrust syphilis down to hell and other re-

joyceana: studies in the border-lands of literature and
medicine. Dublin: Glendale, 1988.

36. Lowbury E. Apollo: an anthology of poems by doctor po-

ets. London: Keynes Press, 1990.

37. Richards lA. Science and poetry. New York: Haskell

House, 1974.

38. Hudson AJ. Literature and medicine: traditions and in-

novations. In: Clarke B, Aycock W, ed. The body and
the text: comparative essays in literature and medi-
cine. Houston: Texas Tech, 1990.

39. Cherubini A, ed. Medici scrittori d'Europa e d'America,

vol. 2. Rome: Delfino, 1990.

40. Charach R. The naked physician: poems about the lives

of patients and doctors. Kingston: Quarry, 1990.

41. Clarke B, Aycock W. The body and the text: comparative

essays in literature and medicine. Lubbock: Texas
Tech, 1990.

42. Poirier S, Borgenicht L. Physician-authors — prophets or

profiteers? N Engl J Med 1991; 325:212-214.

43. Reynolds R, Stone J, eds. On doctoring: stories, poems

essays. New York: Simon and Schuster, 1991.

44. Hunter KM. Doctor's stories: the narrative structure of

medical knowledge. Princeton: Princeton, 1991.

45. Paige NM, Alloggiamento T, eds. Vital signs. The UCLA

collection of physicians' poetry. Los Angeles: UCLA
Medical Center, 1991.

46. Wile IS. The worship of Asklepios, with special reference

to the tholos and the theater. Ann Med Hist 1926;
8:419-434.

47. Hatchett R. Suffering, literature, and the art of medi-

cine. Pharos 1992; 55:13-17.

48. Herring PF. Joyce's uncertainty principle. Princeton:

Princeton, 1987.

49. Hayles NK. Chaos and order: complex dynamics in lit-

erature and science. Chicago: University of Chicago,
1991.

50. Weiss R. Techy to trendy, new products hum DNA's

Vol. 60 No. 2 PHYSICIANS AND LITERATURE— GREEN 153

tune: designers work molecular frequencies into wind
chimes, ink and the song of insulin. New York Times,
September 8, 1992, CI and CIO.

51. Green JP. The application of quantum mechanics to

pharmacology: some substances active in the nervous
system. In: Walaas O, ed. Molecular basis of some as-
pects of mental activity. New York: Academic, 1967;
2:95-111.

52. Wright V. In defence of eponyms. Br Med J 1991; 303:

1600-1602.

53. Asher R. Munchausen's syndrome. Lancet 1951; 1:339-

341.

54. Ifudu O, Kolasinski SL, Friedman EA. Brief report: kid-

ney-related Munchausen's syndrome. N Engl J Med
1992; 327:388-389.

55. Janowitz HD. The art of the gastrointestinal consulta-

tion: some aphorisms. Mt Sinai J Med 1992; 59:341-
349.

56. Sullivan CA, Francis GL, Bain MW, Hartz J. Mun-

chausen syndrome by proxy: 1990 — a portent for prob-
lems? Clin Pediatr 1991; 30:112-116.

57. Dixon B. Lyme and the layperson. Br Med J 1992; 305:

318.

58. Bean WB. The Munchausen syndrome. Perspect Biol

Med 1959; 2:347-353.

59. Eisenberg A. Metaphor in the language of science. Sci

Am 1992; 266:144.

60. Stafford BM. Body criticism: imaging the unseen in En-

lightenment art and medicine. Cambridge: MIT Press,
1991.

61. LaCharit^ RC. Rabelais: the book as therapy. In: Peschel

ER, ed. Medicine and literature. New York: Neale
Watson Academic, 1980; 11-17.

62. Kimmel PL. Francois Rabelais: satire as therapy for the

body politic. Pharos 1992; 55:7-12.

63. Steiner G. Cat man. New Yorker, August 4, 1992, 81-84.

64. Br6e G. The persona of the doctor in Celine's Journey to

the End of the Night and Camus's The Plague. Peschel
ER, ed. Medicine and Literature. New York: Neale
Watson Academic, 1980:88-97.

65. Knapp BL. From hero to horror: Louis-Ferdinand C61ine,

M.D. In: Peschel ER, ed. Medicine and literature. New
York: Neale Watson Academic, 1980; 18-27.

66. Vitoux F. Celine: a biography. Trans, by Browner J. New

York: Paragon House, 1992.

67. Bettmann OL. A pictorial history of medicine. Spring-

field: Charles C Thomas, 1956.

68. Benn G. Selected poems. Wodtke FW, ed. New York: Ox-

ford, 1970.

69. Benn G. Prose, essays, poems. Sander V, ed. The German

Library in 100 Volumes, vol. 73. New York: Contin-
uum, 1987.

70. Esslin M. Review of Tales of a Long Night by Alfred

Doblin. NY Times Book Review, December 30, 1984,
10.

71. Hohmann L, Friedrich Wolf Berlin: Henschelverlag,

1988.

72. Wolf F. Professor Mamlock. New York: Universum,

1935.

73. Bulgakov M. The master and Margarita. Trans, by Gins-

burg M. New York: Grove Weidenfeld, 1967.

74. Bulgakov M. Six Plays. Introduced by Milne L. Trans, by

Powell W, Earley M. London: Methuen, 1991.

75. Bulgakov M. A country doctor's notebook. Trans, by

Glenny M. London: Collins Harvill, 1990.

76. Bulgakov M. Black Snow. Trans, by Glenny M. London:

Harvill, 1986.

77. Bulgakov M. Heart of a Dog. Trans, by Ginsburg M. New

York: Grove, 1968.

78. Clines FX. Satirist gets pedestal (what would Stalin

say?). New York Times, May 16, 1991, A4.

79. Evans Lord. Keats — the man, medicine and poetry. Br

Med J 1969; 3:7-11.

80. Goellnicht DC. The poet-physician: Keats and medical

science. Pittsburgh: University of Pittsburgh, 1984.

81. DeAlmeida H. Romantic medicine and John Keats. New

York: Oxford, 1991.

82. Coope J. Mania Sakhalinosa: an episode in the life of Dr.

Anton Chekhov. Med Hist 1979; 23:29-37.

83. Yarmolinsky A. Letters of Anton Chekhov. New York:

Viking, 1973.

84. Wolff L. The 20th century: Dr. Schnitzler's diagnosis.

New York Times Book Review, November 8, 1992,
1,24-27.

85. Ober WB. A few kind words about W. Somerset

Maugham (1874-1965). NY State J Med 1969; 69:
2392-2701.

86. Key JD. Medical writings of a literary physician: Sir

Arthur Conan Doyle (1859-1930). Minn Med 1978; 61:
362-365.

87. Rodin AE, Key JD. Arthur Conan Doyle's thesis on tabes

dorsalis. JAMA 1982; 247:646-650.

88. Rodin AE, Key JD. Medical casebook of Doctor Arthur

Conan Doyle. Malabar: Robert E. Krieger, 1984.

89. Rodin AE, Key JD. Doctor Arthur Conan Doyle's pa-

tients in fact and fiction. Med Heritage 1985; 1:80-98.

90. Key JD, Rodin AE. Medical reputation and literary cre-

ation: an essay on Arthur Conan Doyle versus Sher-
lock Holmes 1887-1987. Adler Museum Bull 1987; 13:
21-25.

91. Ball D. Sir Arthur Conan Doyle 1859-1930. Practitioner

1975; 215:359-368.

92. Kittle CF. Arthur Conan Doyle: doctor and writer (1859-

1930). J Kansas Med Soc 1960; 61:13-18.

93. Snyder C. There's money in ears, but the eye is a gold

mine; Sir Arthur Conan Doyle's brief career in oph-
thalmology. Arch Ophthal 1971; 85:359-365.

94. Lax E. Woody Allen. New York: Alfred A. Knopf, 1991;

154.

95. Mann RJ, Key JD. Joseph Bell, M.D., F.R.C.S.: "Notes on

a case of paralysis following diphtheria." Pharos 1982;
45:27-29.

96. Scarlett EP. Doctor out of Zebulun: the old original.

Notes on Dr. Joseph Bell whose personality and pecu-
liar abilities suggested the creation of Sherlock
Holmes. Arch Int Med 1964; 114:696-701.

97. Westmoreland BF. Arthur Conan Doyle, Joseph Bell,

and Sherlock Holmes: a neurologic connection. Arch
Neurol 1991; 48:325-329.

98. Belkin BM, Neelon FA. The art of observation: William

Osier and the method of Zadig. Ann Int Med 1992;
116:863-866.

99. Harnagel EE. Doctors afield: Joseph Bell, M.D. — the real

Sherlock Holmes. N Engl J Med 1958; 258:1158-1159.

100. Jackson PW. On learning to see what is not there. Per-

spect Biol Med 1992; 35:499-510.

101. Ober WB. Conan Doyle's dying detective: problem in dif-

ferential diagnosis. NY State J Med 1967; 67:2141-
2145.

102. McDiarmid EW. Foreword. In: Rodin AE, Key JD, eds.

154

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

Medical casebook of Doctor Arthur Conan Doyle: from
practitioner to Sherlock Holmes and beyond. Malabar:
Robert E. Krieger, 1984.

103. Brain P. Dr. Watson's war wounds. Lancet 1969; 2:1354-

1355.

104. Aronson ME. Sherlock Holmes, father of forensic pathol-

ogy. Trans & Studies College of Physicians of Phila-
delphia 1978; 45:258-261.

105. Neblett W. Sherlock's logic. New York: Dorset, 1988.

106. Bates HR. Dr. Watson and the Jezail bullet. Virginia

Med J 1976; 103:828-829.

107. Wolff J. That was no lady: a reply to Mr. Stout in which

are included some observations upon the nature of Dr.
Watson's wound. Am J Surgery 1942; 58:310-312.

108. Rodin AE, Key JD, eds. Conan Doyle's tales of medical

humanism and values: round the red lamp — being
facts and fancies of medical life, with other medical
short stories. Malabar: Krieger, 1992.

109. Brenner RP. Holmes, Watson and neurology. J Clin Psy-

chiatry 1980; 41:202-205.

110. Cherington M. Sherlock Holmes: neurologist? Neurology

1987; 37:824-825.

111. Danek A, Helmchen C, Botzel K. The meaning of Sher-

lock Holmes. Arch Neurol 1992; 49:350.

112. Musto DP. Sherlock Holmes and heredity. JAMA 1966;

196:165-169.

113. Rodin AE. Autoexperimentation with a drug by Arthur

Conan Doyle. J Hist Med Allied Sci 1980; 35:426--430.

114. Anonymous. St. Mary's Hospital. Introductory address

on "The Romance of Medicine," by Sir Arthur Conan
Doyle. Lancet 1910; 2:1066-1068.

115. Doyle AC. The war in South Africa: the epidemic of en-

teric fever at Bloemfontein. Br Med J 1900; 2:49-50.

116. Doyle AC. The Edalji case. Br Med J 1907; 1:173.

117. Clarity JF. What's doing in Dublin. New York Times,

June 21, 1992, sect. 5, 10.

118. Fischoff E. Oliver Wendell Holmes: physician and hu-

manist. Pearson Museum Monograph Series 1982; 82/
2:1-17.

119. Williams WC. The doctor stories, compiled with an in-

troduction by Coles R, afterword by W. E. Williams.
New York: New Directions, 1984.

120. Williams WC. The autobiography of William Carlos

Williams. New York: New Directions, 1967.

121. Holub M. Selected poems. Translated by Milner I,

Theiner G. Middlesex: Penguin, 1967.

122. Sanger D. A Noh drama poses new question in old forms.

New York Times, April 1, 1991, Cll.

123. Papper EM. The influence of romantic literature on the

medical understanding of pain and suffering — the
stimulus to the discovery of anesthesia. Perspect Biol
Med 1992; 35:401^15.

124. Leithauser B. Voices in the clock. New Yorker, July 8,

1991, 80-84.

125. Timpane J. Essay: the poetry of science. Sci Am 1991;

265:128.

126. Medawar P. Pluto's republic. New York: Oxford, 1984.

127. Sacks O. Foreward to Luria AR, The history of a brain

wound: the man with a shattered world. Cambridge:
Harvard, 1987.

128. Spiro H. What is empathy and can it be taught? Ann Int

Med 1992; 116:843-846.

129. Oglesby P. The caring physician: the life of Dr. Francis

W. Peabody. Boston: The Francis A Countway Library
of Medicine, 1991.

130. Baron RJ. Bridging clinical distance: an empathic redis-

covery of the known. J Med Philos 1981; 6:5-23.

131. Baron RJ. An introduction to medical phenomenology: I

can't hear you while I'm listening. Ann Int Med 1985;
103:606-611.

132. Eisenberg L. Mindlessness and brainlessness in psychi-

atry. Br J Psych 1986; 148:497-508.

133. Eisenberg L. Science in medicine: too much or too little

and too limited in scope? Am J Med 1988; 84:483^91.

134. Brody S. We have lost our humanity. Newsweek, Sep-

tember 7, 1992, 8.

135. Breo DL. Uwe Reinhardt, PhD — the economist as health

evangelist. JAMA 1992; 268:1332-1336.

136. Belkin L. Hospital study tests benefits of giving comfort

with care. New York Times, Sept. 26, 1992, 1,24.

137. Mai FM. New diagnostic instrument brings psyche intc

view. Ontario Med Rev 1983; 50:515-517.

138. Haggerty RJ, Bloom SW, Mechanic D, Pardes H. Report

of the commission's subcommittee on the behavioral
sciences. In: Marston RQ, Jones RM. Medical educa
tion in transition. Commission on medical education
the sciences of medical practice. Princeton: Robert
Wood Johnson Foundation, 1992:74-78.

139. Odegaard CE, Inui TS. A 1992 manifesto for primar)

physicians. Pharos 1992; 55:2—6.

140. Weisberg R. Poethics and other strategies of law anc

literature. New York: Columbia University, 1991.

141. Coles R. Book value. Dartmouth Medicine 1991; 15:18-

24,53.

142. Trautmann J. The wonders of literature in medical edu

cation. Mobius 1982; 2:23-31.

143. Gogel EL, Terry JS. Medicine as interpretation: the uses

of literary metaphors and methods. J Med Philos 1987
12:205-217.

144. Cassell EJ. The place of the humanities in medicine

Hastings-on-Hudson: The Hastings Center, 1984.

145. Bellet PS, Maloney MJ. The importance of empathy at

an interviewing skill in medicine. JAMA 1991; 266
1831-1833.

146. Epstein J. Historiography, diagnosis, and poetics. Lii

Med 1992; 11:23-44.

147. Belkin L. In lessons on empathy, doctors become pa

tients. New York Times, June 4, 1992, Al, B5.

148. Cousins N, ed. The physician in literature. Philadelphia

Saunders, 1982.

149. Radey C. Telling stories: creative literature and ethics

Hastings Center Report 1990; (November/December)
25.

150. Sheriff DS. Literature and medical ethics. J Roy Soc Met

1988; 81:688-690.

151. Streeter RE. The case of Tertius Lydgate. Persp Biol Mec

1989; 32:171-173.

152. Donnelly WJ. Experiencing the death of Ivan Ilyich: nar

rative art in the mainstream of medical education
Pharos 1991; 54:21-25.

153. Radwany SM, Adelson BH. The use of literary classics ir

teaching medical ethics to physicians. JAMA 1987
257:1629-1631.

154. Lifshitz LH, ed. Her soul beneath the bone: women's po

etry on breast cancer. Urbana: University of Illinois
1988.

155. Wear D, Nixon LL. "Scoot down to the edge of the table

hon": women's medical experiences portrayed in liter
ature. Pharos 1991; 54:7-11.

156. Secundy MG, ed., with the literary collaboration o

Vol. 60 No. 2

PHYSICIANS AND LITERATURE— GREEN

155

Nixon LL. Trials, tribulations, and celebrations: Afri-
can-American perspectives on health, illness, aging
and loss. Yarmouth, Maine: Intercultural Press, 1992.

157. Schama S. Dead certainties (unwarranted speculations).

New York: Knopf, 1990.

158. Hjort PF. Is high technology medicine a threat to health

care? Ann Med 1992; 24:145-147.

159. Peterson MC, Holbrook JH, Hales DV, Smith NL, Staker

LV. Contributions of the history, physical examina-
I tion, and laboratory investigation in making medical

diagnoses. West J Med 1992; 156:163-165.

160. Broyard A. Intoxicated by my illness: and other writings

on life and death. New York: Clarkson Potter, 1992.

161. Feher M, NaddaffR, Tazi N, eds. Fragments for a history

of the human body, parts 1, 2, 3. Cambridge: MIT
Press, 1989.

162. Bertman SL. Facing death: images, insights, and inter-

ventions: a handbook for educators, healthcare profes-
sionals, and counselors. New York: Hemisphere Pub-
lishing, 1991.

163. Winkler MG. Seeing patients. Lit Med 1992; 11:216-222.

164. Blum A. Sketches. Lit Med 1992; 11:223-236.

165. Magarshack D. Introduction. In: Dostoyevsky F. The id-

iot. Baltimore: Penguin, 1958.

166. Kael P. The current cinema: the doctor and the director.

The New Yorker, February 11, 1991, 70.

Prevalence of Thyroid Autoantibodies in
Ambulatory Elderly Women

Carol Martinez- Weber, M.D., Priscilla F. Wallack, M.D., Pearl Lefkowitz, M.D., and

Terry F. Da vies, M.D.

Abstract

Major improvements in thyroid autoantibody testing have now become widely available.
Seventy-five elderly ambulatory women fi"om a senior citizens' center (mean age 74.5 yrs)
were studied to reassess the prevalence of thyroid autoantibodies and to demonstrate how
such tests related to clinical signs and symptoms of thjrroid disease. We used the enzyme-
linked immunoassay (ELISA) technique for the measurement of autoantibodies to thy-
roglobulin (hTg) and thyroid peroxidase (hTPO) (as microsomal antigen), since ELISA
systems are economical, highly sensitive and specific for population screening. Autoanti-
bodies to hTPO (hTPO-Ab) were present in 44% and hTg autoantibodies (hTg-Ab) in 32%
of the study group. Ten women (13.3%) had elevated thyrotropin (TSH) levels. An elevated
serum TSH was associated with the presence of hTPO-Ab in varying concentrations. The
mean TSH value of 7.2 jjiIU/mL in those women with hTPO-Ab was significantly higher
than the mean of 4.7 ixIU/mL found in those women without thyroid autoantibodies {p <
0.01). However, additional testing for hTg-Ab was of little clinical value.

These data indicate the high prevalence of thyroid autoimmune disease in the elderly
female population. We conclude that screening for thyroid dysfunction is best achieved by
the measurement of serum TSH in all women over the age of 60 years. The measurement
of hTPO-Ab, but not hTg-Ab, was helpful in confirming the cause of thjToid failure in the
elderly female population.

Thyroid disease in the elderly may present insid-
iously with nonspecific signs and symptoms erro-
neously attributed to the normal aging process or
other medical conditions. This is particularly so
for thyroid failure secondary to atrophic thyroid-
itis, where the absence of a goiter may mask the
clinical diagnosis. Difficulty in the clinical diag-
nosis of mild thyroid failure has led to the sug-
gestion that routine biochemical screening for
thyroid disease in the elderly may be of value. An
elevated thyrotropin (TSH) level remains the
most sensitive indicator of thyroid deficiency (1,
2) and in various clinical studies, elevated TSH
values have been found in 2.6% to 24% (3-5) of

From the Departments of Community Medicine and Medicine,
St. Vincent's Hospital and Medical Center, and Department of
Medicine (TFD), Mount Sinai School of Medicine, New York,
NY. Address correspondence and reprint requests to Dr. T. F.
Davies, Box 1055, Mount Sinai Medical Center, One Gustave
L. Levy Place, New York, NY 10029.

the elderly. Autoantibodies to human thyroid per-
oxidase (the microsomal antigen) (hTPO-Ab) and/
or human thyroglobulin (hTg-Ab) autoantibodies
have also been found in 7.4% to 42.2% of the el-
derly (2, 5-8). Unfortunately, these reports vary
in patient characteristics, setting, and assay
methodology, thus making it difficult to draw
broad conclusions. However, most show a predom-
inance of increased TSH levels and of hTPO-Ab
and hTg-Ab in elderly women. The recent intro-
duction of more sensitive techniques for the mea-
surement of thyroid function and thyroid au-
toantibodies has caused such studies to require
ree valuation, since it is now easier to apply them
on a large scale. We have, therefore, performed
a cross-sectional study of elderly, ambulatory
women to determine the prevalence of abnormal
thyroid function tests, including hTPO-Ab and
hTg-Ab, by ELISA technology. This study has
also allowed us to understand the relationship
between such thyroid function tests and the

156

The Mount Sinai Journal of Medicine Vol 60 No. 2 March 1993

Vol. 60 No. 2

THYROID IN ELDERLY WOMEN— MARTINEZ-WEBER ET AL.

157

signs and symptoms of thyroid disease in the el-
derly.

Methods

Patient Selection. Through an outreach pro-
gram of the Department of Community Medicine,
St. Vincent's Hospital and Medical Center of New
York, volunteers were recruited at a senior citi-
zens' center in the Chelsea neighborhood of New
York City. There were approximately 1100 older
people enrolled at the center, of whom 650 at-
tended regularly, lived independently in the com-
munity, and traveled to the center for social con-
tact. Three-quarters of the members were women.
Seventy-five percent of the center members were
white, 18%-20% Hispanic, and less than 5%
black. All other ethnicities comprised less than
1%. The mean age of the center population was 74
years, with a range from 60 to 90 years. At the
time of interview and examination, none of the
participants were acutely ill, grossly malnour-
ished, or taking any medications, other than thy-
roid replacement, known to interfere with the
measurement of thyroid function.

Protocol. For a 15-month period, the oppor-
tunity for thyroid screening was offered to all cen-
ter attendees through announcements at monthly
meetings and a newsletter. After the purpose and
protocol of the study were explained to the volun-
teers, informed consent was obtained. The sub-
jects were evaluated using a questionnaire, phys-
ical examination, and laboratory studies which
focused on data relevant to thyroid disease. The
standard signs and symptoms of thyroid disease,
as defined in general medical texts, were sought.
Each participant was screened by one of two phy-
sician examiners who completed all aspects of the
procedure in one session at the senior citizens'
center. Appropriate follow-up was arranged for
any abnormalities detected.

Assays. Laboratory studies included serum
total T4, total T3, T3 resin uptake, thyrotropin
(TSH), and thyroid peroxidase (hTPO) and thyro-
globulin (hTg) autoantibodies (Ab). Specific im-
munoassays were used for T4 (T4-Tetra-Tab RIA
by Nuclear Medical Labs), T3 (Abbott Labs), T3
resin uptake (Tri-Tab T3 Uptake by Nuclear
Medical Labs), and TSH (Thyro-Shure by Nuclear
Medical Labs) (normal range 0.5—6 ixIU/mL).
Thyroid peroxidase (as microsomal antigen) and
thyroglobulin autoantibodies were measured by
enzyme-linked immunosorbent assays (9) with re-
sults expressed in terms of an ELISA Index. Pa-
tients' results were obtained by dividing the ab-
sorbance of the unknown serum sample at a 1:100

dilution by the absorbance of a 10,000-fold dilu-
tion of appropriate standard. Background binding
was removed from both data sets. Use of the index
allows comparison between assays (9). The nor-
mal range for the ELISA indices, based on normal
control population studies, was less than 0.2. We
have found the ELISA technique to be 17% more
sensitive than hemagglutination for assay of hTg-
Ab and 12% more sensitive for hTPO-Ab (9).

Analysis. Data analysis was performed using
the CLINFO statistical package of the Clinical
Research Center at the Mount Sinai School of
Medicine. The functions utilized included the
Fisher exact test, chi-square analysis, and Stu-
dent's ^-test with a pre-set significant P value of
<0.05.

Results

Demographics and Symptoms and Signs.

The first 75 women volunteers aged 61-87 years
were surveyed. Their mean age was 74.5 years.
Ninety-two percent were white, 5.3% Hispanic,
1.3% black, and 1.3% native American. Twenty-
three of the 75 (31%) had a known history of thy-
roid disease. In the review of systems, the preva-
lence of certain nonspecific symptoms of
hypothyroidism was: memory loss, 50 subjects
(67%); weakness, fatigue, or lethargy, 27 (36%);
depression, 22 (29%); constipation, 17 (23%). Bra-
dycardia was not detected. A goiter was present
in 13 participants (17%), three of whom had ab-
normal TSH values (see below), and another four
of whom had a history of thyroid replacement
therapy more than 10 years earlier. Clinical signs
of overt hypothyroidism were found in only one
subject (1.3%).

Thyroid Function Tests. The mean T4, T3,
and T3 resin uptake were 8.8 |JLg/dL, 117.7 mg/dL,
and 37.5 respectively. The median TSH value was
4.5 fxIU/mL; the mean TSH value of 5.8 M-IU/mL
was skewed to the right by the presence of three
high values (see the Fig.). Only one woman (1.3%)
in the study had an abnormally low T4 value.
This participant had no known history of thyroid
disease, but was clinically hypothyroid, with a
markedly increased TSH. Ten of 75 participants
(13.3%) had an elevated TSH defined as greater
than 7 jxIU/mL (two standard deviations above
the laboratory mean) (Table 1). Three of these 10
women had a history of thyroid disease but were
not receiving treatment; seven were newly diag-
nosed. There was no significant association found
between an increased TSH level and patient de-
mographics, signs and symptoms of thyroid dis-
ease, or T4, T3, and T3 resin uptake values (Ta-
ble 1).

158

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

TSH (>ilU/ml)

Fig. Comparison between serum TSH levels ((xIU/mL) and
the level of hTPO-Ab measured as ELISA absorbance.

Thyroid Autoantibodies. Autoantibodies to
hTPO were present in 33 of the 75 women (44%),
and to hTg-Ab in 24 subjects (32%) (Table 2).
Both autoantibodies were found in 21 subjects
(28%), and 36 subjects (48%) had one or both au-
toantibodies present. There was a direct correla-
tion between the level of hTPO-Ab and the level
of hTg-Ab; those with high titers of one antibody
tended to have high titers of the other (r = 0.633,
p = 0.001). Of the 13 participants with a goiter, 8
(62%) had hTPO-Ab present, and 6 (45%) had
hTg-Ab present. None of those with a goiter had
hTg-Ab without hTPO-Ab.

Nine of the 10 subjects (90%) with TSH val-
ues greater than 7 |xIU/mL had detectable hTPO-
Ab and five (50%) had hTg-Ab. Of the 65 partic-
ipants with normal TSH values, 24 (37%) had
hTPO-Ab present and 19 (29%) had hTg-Ab. An
elevated TSH was, therefore, significantly associ-
ated with the presence of hTPO-Ab (Fisher exact
test p < 0.005). The mean TSH in those patients
with hTPO-Ab present was 7.2 ixIU/mL compared
to 4.7 in those without hTPO-Ab. This difference
was statistically significant by two-tailed Stu-
dent's t-test (P < 0.02) (Table 3). However, the
titer of hTPO-Ab did not correlate with the TSH
level (Fig.). No independent association was
found between TSH and hTg-Ab.

There was no significant association between
the presence or absence of either thyroid au-
toantibody and patient demographics, signs and
symptoms, or other thyroid function tests, T4, T3,
or T3 resin uptake values, based on chi-square
analyses.

Goiter Patients. Thirteen patients had a clin-
ically detectable goiter including two with palpa-
ble nodules. Three patients had increased serum
TSH and two additional patients were receiving
thyroxine replacement therapy, indicating that 5

of the 13 had significant thyroid failure (38%). Six
of the goiter patients had one or the other thyroid
autoantibody detectable, and these included the 5
patients with thyroid deficiency.

Discussion

As could be expected in an elderly popula-
tion, symptoms of thyroid hypofunction such as
memory loss, fatigue, depression, and constipa-
tion were prevalent but probably nonspecific. A
goiter was present in 13 subjects, 8 of whom had
never taken thyroid medication and who were
clinically and biochemically euthyroid. Only one
participant in the study group of 75 (1.3%) had
symptoms and signs of over hypothyroidism.

Our results show a 13.3% prevalence of ele-
vated TSH levels in ambulatory women over the
age of 60 years. This finding substantiates those
of Sawin et al. (4) and of Tunbridge et al. (3) and
a large recent cross-sectional study of 1200 elder-
ly patients (10). Both found a high prevalence of
elevated TSH values in elderly women, 24% and
17.5%, respectively. These studies also noted a
significantly lower prevalence rate in elderly
men. Sawin et al. used the TSH upper limit of 5
jxIU/mL to define an elevated TSH, which may
explain the higher prevalence rates in his study.
A New Zealand study which screened elderly per-
sons over the age of 80 years using total T4 levels
found 0.94% below normal (11). More cases of sub-
clinical hypothyroidism might have been detected
if TSH values had been examined. Since only one
of ten women in our study with an elevated TSH
level had a low T4 (10%), TSH is a far more sen-
sitive screening test for hypothyroidism than T4.
However, a prevalence of increased TSH values of
only 2.6% was found in a previous United States
study that did screen adults over 60 years of age
with TSH (5). These authors did not indicate the
ratio of men to women. As the prevalence of ele-
vated TSH in men is lower than in women be-
cause of the lower prevalence of autoimmune thy-
roid disease in men, a male predominance in the
sample may have contributed to this lower rate.

A more recent study in the United States (12)
and several from other countries (3, 6, 7, 8, 10)
document that 7%-45% of elderly subjects had
hTPO-Ab or hTg-Ab or both. However, a variety
of antibody techniques were used in these studies.
Among surveys utilizing the same assays, the
chosen upper limit of normal has varied, thus fur-
ther limiting comparisons. In our study, the pres-
ence of thyroid autoantibody constituted a posi-
tive test. We found a 44% prevalence of hTPO-Ab
in ambulatory women over the age of 60 years. In

Vol. 60 No. 2

THYROID IN ELDERLY WOMEN— MARTINEZ-WEBER ET AL.

159

TABLE 1

TSH versus Age and Thyroid Function Studies

TSH < 7 (n = 65) TSH > 7 (n = 10) Total

Mean

SD*

Range

Mean

SD*

Range

Mean ± SD

Range

Age (years)

74.9

7.7

61-87

71.8

5.5

65-83

74.5 ± 6.6

61-87

T4 (fig/dL)

9.1

1.6

4.9-13.5

2.0

2.2-9.7

8.8 ± 1.8

2.2-13.5

T3 (mg/dL)

118.5

28.2

75.8-255.1

112.7

23.4

74.1-145.9

117.7 ± 27.5

74.1-255.1

T3RU (%)

37.5

2.8

32.8-45.2

36.8

2.9

33.0-41.8

37.5 ± 2.8

32.8-45.2

TSH (M-IU/mL)

5.8 ± 4.6

1.0-28.6

* No significant difference between these two groups by Student's ^test.

the only other prevalence study in the elderly uti-
lizing ELISAs (6), a rate of 18.5% was found,
much lower than our 44% in a similar elderly
female population. This study limited the positive
hTPO-Abs to those participants with antibody ti-
ters greater than expected in an elderly control
population. Since there is a higher prevalence of
hTPO-Ab in the elderly, this population may not
have been a suitable control group (6).

Our data support (8, 12) a significant associ-
ation between an elevated TSH value and the
presence of hTPO-Ab. However, our study and
others (8, 10, 11) reveal that one-quarter to one-
half of patients with hTPO-Ab present have nor-
mal TSH levels. Figure 1 illustrates the high an-
tibody titers present in many subjects with
normal TSH levels. Data are available suggesting
that individuals with raised TSH and thyroid au-
toantibodies are at higher risk of developing thy-
roid failure (13), but the prognosis for those with
high antibody titers and normal TSH levels has
been less well documented. However, 90% of our
subjects with subclinical thyroid failure (as evi-
denced by increased TSH levels) had detectable
hTPO-Ab. Using a less sensitive antibody assay,
another study (12) found a 67% prevalence of
hTPO-Ab among persons with increased TSH val-
ues.

Hence, the clinical importance of hTPO-Ab in
the presence of a normal TSH level and normal

TABLE 2

TSH versus Thyroid Antibodies

TSH TSH

< 7 (jiIU/mL

7 jjLlU/mL

Total

n %

hTPO-Ab

present

(37)

(90)

33 (44)

absent

(63)

(10)

42 (56)

hTg-Ab

present

(29)

(50)

24 (32)

absent

(71)

(50)

51 (68)

thyroid function still remains to be established
but may be related to the subclass of the antibod-
ies present (14). At present the only known cor-
relation is with subclinical autoimmune thyroid
disease, with evidence of mononuclear thyroid in-
filtration, the degree of infiltration correlating
with antibody titer. Of interest, however, are ob-
servations that age itself is not directly related to
the onset of thyroidal lymphocytic infiltration
and that racial background may be of great im-
port (15). Recent research suggests a link be-
tween characteristic thyroid autoantibody immu-
noglobulin G subclass distributions and titers and
an inherited predisposition to thyroid disease
(14). In the future, these may serve as markers for
people at risk for developing clinical thyroid fail-
ure. However, women with an elevated TSH and
the presence of thyroid hTPO-Ab have been found
to have a 5% per year development rate of overt
hypothyroidism (13). This development was cu-
mulative, so that over a four-year period 20%
were affected, and it was projected that over 10
years, 50% might develop clinical hypothyroid-
ism. Further prospective studies are necessary to
confirm such predictions in the elderly popula-
tion, but one recent study followed such patients
up for 12 months and found 18% of patients re-
quired thyroxine therapy during this time period
(10).

While low TSH values in the elderly are often
misleading and cannot be considered diagnostic of
thyroid overactivity (10), the question of whether
or not to treat an elevated TSH level in clinically
euthyroid individuals has attracted much atten-
tion and is of major import in the elderly popula-
tion, where coincident disease is more likely to be
present. One retrospective study (16) found that
normalization of an elevated TSH correlated with
resolution of mild nonspecific symptoms of h5rpo-
thyroidism, such as dry skin and easy fatigabil-
ity. However, patients with a history of ablative
thyroid therapy were studied, so that the results
cannot be generalized to include other forms of

160 THE MOUNT SINAI JOURNAL OF MEDICINE March 1993

TABLE 3

Thyroid Antibodies versus Thyroid Function Studies

hTPO-Ab hTg-Ab

Present
(X ± SD)

Absent
(x ± SD)

Present
(x ± SD)

Absent
(x ± SD)

mean T4

((ig/dL)

8.8 ± 2.2

8.8 ± 1.4

8.7 ± 2.4

8.8 ± 1.4

mean T3

(mg/dL)

118.3 ± 33.4

117.2 ± 22.3

121.6 ± 35.5

115.9 ± 23.1

mean TSH

(jilU/mL)

7.2 ± 6.4*

4.7 ± 1.7*

6.8 ± 6.8

5.3 ± 2.9

hTPO-Ab, thyroid peroxidase antibody; hTg-Ab, thyroglobulin antibody; TSH, thyrotropin.
* P < 0.02 by two-tailed Student's i-test.

h3T)othyroidism. More recent studies of cardiac
function in similar patients showed marked im-
provement after thyroid hormone replacement
(17). Since mild congestive heart failure is com-
mon in the elderly population, the treatment of
subclinical thyroid failure may be of considerable
importance.

In conclusion, our study of 75 women be-
tween the ages of 61 and 87 years revealed a prev-
alence of 13.3% of subclinical hypothyroidism, as
evidenced by an elevated TSH level. Forty-four
percent of the participants had hTPO-Ab and 32%
had hTg-Ab. An elevated TSH was significantly
associated with the presence of hTPO-Ab. No
other significant relationships between the symp-
toms and signs of thyroid disease, thyroid func-
tion studies, or the presence of autoantibodies
were found. We recommend screening of women
over the age of 60 years for thyroid dysfunction,
since the high prevalence of disease is not clini-
cally apparent and may contribute to other major
pathology such as cardiac dysfunction. A serum
TSH level is the most sensitive and specific
screening test available at this time and may be
effectively linked with hTPO-Ab assessment to
determine etiology of the thyroid failure.

Acknowledgments

We thank Iven Young M.D., Elsa Echemendia M.D., and
Sheila Roman M.D. for their help and support and Michael
Platzer for technical assistance. The CLINFO statistical pack-
age is supported by grant RR-71 from NIH-DRR to the Clin-
ical Research Center.

References

1. Evered DC, Ormston BJ, Smith PA, et al. Grades of hy-

pothyroidism. Br Med J 1973; 1:657-662.

2. Tibaldi J, Barzel US. Thyroxine supplementation: method

for prevention of clinical hypothyroidism. AJM 1985;
79:241-244.

3. Tunbridge WMG, Evered DC, Hall R, et al. The spectrum

of th5Toid disease in a community: the Whickham Sur-
vey. Clin Endocrinol 1977; 7:481-493.

4. Sawin CT, Chopra D, Azizi F, et al. The aging thyroid:

increased prevalence of elevated serum thyrotropin lev-
els in the elderly. JAMA 1979; 3:247-250.

5. Heikoff LE, Luxenberg J, Feigenbaum LZ. Low yield of

screening for hypothyroidism in healthy elderly. J Am
Geriatr Soc 1984; 32:616-617.

6. Lazarus JH, Burr ML, McGregor AM, et al. The preva-

lence and progression of autoimmune thjnroid disease in
the elderly. Acta Endocrinol 1984; 106:199-202.

7. Hawkins BR, Dawkins RL, Burger HR, et al. Diagnostic

significance of thyroid microsomal antibodies in a ran-
domly selected population. Lancet 1980; 2:1057-1059.

8. Gordin A, Maatela J, Miettinen A, et al. Serum thyrotro-

pin and circulating thyroglobulin and thyroid microso-
mal antibodies in a Finnish population. Acta Endo-
crinol 1979; 90:33-42.

9. Roman SH, Kom F, Davies TF. Enzyme-linked immuno-

sorbent microassay and hemagglutination compared for
detection of thyroglobulin and thyroid microsomal au-
toantibodies. Clin Chem 1984; 30:246-251.

10. Parle JV, Franklyn JA, Cross KW, Jones SC, Sheppard

MC. Prevalence and follow-up of abnormal TSH concen-
trations in the elderly in the United Kingdom. Clin
Endocrinol 1991; 34:77-83.

11. Campbell AJ, Reinken J, Allan BC. Thyroid disease in the

elderly community. Age Aging 1981; 10:47-52.

12. Sawin CT, Bigos ST, Land S, Bacharach P. The aging

thyroid: relationship between elevated serum thyrotro-
pin levels and thyroid antibodies in elderly patients.
Am J Med 1985; 79:591-595.

13. Tunbridge WMG, Brewis M, French JM. Natural history

of autoimmune thyroiditis. Br Med J 1981; 282:258-
262.

14. Davies TF, Weber CM, Wallack P, Platzer M. Restricted

heterogeneity and T cell dependence of human thyroid
autoantibody immunoglobulin G subclasses. J Clin En-
docrinol Metabol 1985; 62:945-949.

15. Okayasu I, Hatakeyama S, Tanaka Y, Sakurai T, Hoshi

K, Lewis PD. Is focal chronic thyroiditis an age-related
disease? Differences in incidence and severity between
Japanese and British. J Pathol 1991; 163:257-264.

16. Cooper DS, Halpern R, Wood LC. L-thyroxine therapy in

subclinical hypothyroidism. Ann Int Med 1984; 101:18-
24.

17. Forfar JC, Wathen CG, Todd WTA, Bell GM, Hannan WJ,

Muir AL, Toft AL. Left ventricular performance in sub-
clinical hypothyroidism. Q J Med 1985; 57:857-865.

MRI Documentation of Hemorrhage into
Post-traumatic Subdural Hygroma

John O. Lusins, M.D. and Ernest R. Levy, M.D.
Abstract

Rapid enlargement of a post-traumatic hygroma was demonstrated by MRI in a patient
who had sequential studies performed after head injury. The enlargement was shown to be
due to hemorrhage into the hygroma.

Accumulation of cerebral spinal fluid between
the dura and arachnoid is a complication which
may arise after head injury. Terms such as men-
ingitis serosa traumatica, traumatic subdural ef-
fusion, and subdural hydroma also have been
used to describe this condition. The most com-
monly accepted one is subdural hygroma.

The first clinical case to be described was
that by Mayo in 1893 (1). NafTziger in 1924 (2)
was the first to postulate that the cause of the
excessive accumulation of the subdural CSF was
due to a tear in the arachnoid at the time of the
trauma. He proposed that the arachnoid tear
acted as one way flap valve permitting CSF to
enter but not leave the space. Therefore, the fluid
would collect and the space enlarge. Gradual en-
largement of a hygroma has been documented by
both CT and angiography (3-6).

Recently, we had the opportunity to evaluate
a patient with a post traumatic hygroma in whom
the enlargement was not solely due to this flap
valve postulate but rather due to bleeding into
the hygroma collection.

Case Report. A 74-year-old man lost control
of his vehicle and struck his head against the
windshield, losing consciousness. He regained

From the Otsego Magnetic Resonance Imaging Center, One-
onta, NY, Dept. of Neurology, Mount Sinai School of Medicine,
New York City, Dept. of Neurosurgery, Mary Imogene Bassett
Hospital, Cooperstown, NY, and Dept. of Neurosurgery, Co-
lumbia Presbyterian School of Medicine, New York City.

Address reprint requests to Dr. John O. Lusins, 395 Main
Street, Oneonta, NY 13820.

consciousness in the Emergency Room and re-
turned to full mental function within two hours
after the accident. A large laceration was noted
on the back of the head.

The patient was admitted to the hospital for
observation. On admission, the neurological ex-
amination failed to show any lateralizing find-
ings. The patient had a CT scan which demon-
strated the contusion in the occipital area and
demonstrated a small but distinct frontal hyper-
lucency overlying the hemispheres and diagnosed
as a left frontal hygroma (Fig. 1). The patient
was followed up as an outpatient. The patient was
seen a week later. At that time, the neurologic
examination was negative and his mental status
was normal. No headache was reported. A repeat
MRI showed the hygroma collection to be present
and only minimally changed from the previous
study (Fig. 2).

It was elected to continue to observe the pa-
tient. The patient was seen two weeks later, at
which time he was complaining of developing left-
sided headache and some difficulty in concentrat-
ing. No lateralizing findings were noted. MRI
showed significant enlargement of the subdural
fluid collection, the fluid now being hyperintense
on Tl and with a meniscus effect. A significant
midline shift was noted (Fig. 3). The patient un-
derwent surgery, and hemorrhagic fluid collec-
tion was removed. No membranes were noted. His
postoperative course was uneventful.

Discussion. Although the first demonstra-
tions of enlargement of a hygroma were through
angiography, not until CT was readily available

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 1993

161

162

THE MOUNT SINAI JOURNAL OF MEDICINE

March 1993

Fig. 1. Left CT obtained at time of injury shows left frontal
hygroma in association with right occipital scalp contusion.
Fig. 2. Right MRI sagittal Tl image performed a week af-
ter initial CT demonstrates previously diagnosed hygroma.

was a significant series collected. In the largest
CT series by French et al. (5), the incidence of
post-traumatic hygromas was 6.6% in a consecu-
tive series of 1,601 patients with craniocerebral
trauma. In that series, 54% of the cases (7 of 13)
initially had a normal CT, only later developing
the hygroma.

Since Naffziger's initial description of a tear
in the arachnoid, numerous published reports
have dealt with post-traumatic subdural hy-
groma. However, only in a relatively few in-
stances have the authors been able to demon-
strate at surgery an actual tear in the arachnoid
(6, 7). Radionuclear studies, however, do lend cre-
dence to this postulate, since radioactively tagged
albumen injected by lumbar puncture has been
shown to gradually accumulate in the hygroma
collection (6, 8).

In reviewing the literature, we are unable to

^^^^^^^^^^^^^^^^^^^^^^^^^

Fig. 3

MRI (Left) Sagittal and (Right) transaxial Tl
images demonstrate enlargement due to hemorrhage into hy-
groma. Significant midline shift is present.

demonstrate a case in which the hygroma is
thought to enlarge by bleeding rather than by
gradual spinal-fluid entrapment. However, a
number of papers indicate that at surgery there is
a spectrum of findings from purely clear cerebro-
spinal fluid in the hygroma to fluid that is xan-
thochromic or actually blood tinged (7, 9, 10). The
phenomenon of enlargement by bleeding, how-
ever, has been demonstrated to occur in the case
of sudden enlargement of chronic subdural he-
matoma (11). It is likely that both mechanisms
may play a part in enlargement of a hygroma. As
the entrapped CSF gradually enlarges the subdu-
ral space, the bridging veins are increasingly
stretched and in certain cases may rupture, pro-
ducing rapid enlargement of the hygroma.

References

1. Mayo CH. A brain cyst: the results of injury causing apha-

sia, hemiplegia, etc. Evacuation: complete recovery. NY
Med J 1894; 59:434-436.

2. Naffziger HC. Subdural fluid accumulations following

head injury. JAMA 1924; 82:1751-1752.

3. Winestock DP, Spetzler RF, Hoff JT. Acute, posttraumatic

subdural hygroma. Natural course with angiographic
documentation. Radiology 1975; 115:373-375.

4. Jaeckle KA, Allen JH. Subdural hygroma: diagnosis with

computed tomography. J Comput Assist Tomogr 1979;
3:201-206.

5. French BN, Dubin AB. The value of CT in the manage-

ment of 1000 consecutive head injuries. Surg Neurol
1977; 7:171-183.

6. Hoff J, Bales E, Barnes B, Glinchkman M, Margolis T.

Traumatic subdural hygroma. Trauma 1973; 13:870-
876.

7. St. John JN, Dila C. Traumatic subdural hygroma in

adults. Neurosurgery 1981; 9:621-626.

8. So SK, Ogawa T, Gerberg E, Sakimura I, Wright W.

Tracer accumulation in a subdural hygroma: case re-
port. J Nucl Med 1976; 17:119-121.

9. Stone JL, Lang RG, Sugar O, Moody RA. Traumatic sub-

dural hygroma. Neurosurgery 1981; 8:542-550.

10. Oka H, Montomochi M, Suzuki Y, Ando K. Subdural hy-

groma after head injury. Acta Neurochir (Wien) 1972;
26:265-273.

11. Lusins J, Danielski EF, Kreps S. Uncommon CT patterns

of subdural hematoma managed nonsurgically. Mt
Sinai J Med 1988; 55:278-281.

Submitted for publication September 1992.

A Case Report of Neurologic
Improvement Following Treatment of
Paraneoplastic Cerebellar Degeneration

Betty Jane Mintz, M.D., and David K. Sirota, M.D.

Paraneoplastic cerebellar degeneration (PCD)
is a rare remote effect of cancer characterized by
acute or subacute cerebellar dysfunction. Clini-
cally, PCD is a syndrome of truncal and limb
ataxia, sometimes associated with dysarthria and
nystagmus. It occurs in less than 1% of patients
with cancer (1). The onset of PCD may occur be-
fore or after the detection of a systemic malig-
nancy. Pathologically, a profound loss of Purkinje
cells is seen. In some patients with this disorder,
an anti-Purkinje cell autoantibody has been
found, termed anti-Yo (2, 3).

Seropositive and seronegative patients differ
in regard to sex and type of malignancy (4, 5).
Most seropositive patients are postmenopausal
women with breast or gynecologic tumors. Sero-
negative patients can be male or female and have
a wider range of tumors, including breast and gy-
necologic tumors, as well as Hodgkin's disease
and lung carcinoma.

We report on a patient with this uncommon
disorder who was somewhat atypical in several
respects, including her response to treatment.

Case Report. A 70-year-old white woman had
an enlarged, nontender right inguinal lymph
node. Biopsy revealed metastatic adenocarci-
noma. Further study showed enlargement of the
left ovary.

The day after learning her diagnosis, she
complained of sudden inability to control her legs.
On neurologic examination, she was unable to
stand, with or without assistance. There was no

From the Departments of Neurology (BJM) and Internal Med-
icine (DKS), Mount Sinai Medical Center, New York. Address
reprint requests to Betty Jane Mintz, M.D., 1421 Third Ave-
nue, New York, NY 10028.

nystagmus. Rapid alternating movements, fin-
ger-to-nose, and heel-to-shin testing were normal.
Vibratory sensation was diminished in the toes.
The remainder of the examination results were
normal, and the possibility of a conversion reac-
tion was considered in the differential diagnosis.
However, the next day she developed evidence of
cerebellar dysfunction with marked impairment
in finger-to-nose and heel-to-shin testing. She re-
mained unable to walk even with assistance.
Magnetic resonance imaging with contrast
showed only minimal cerebral atrophy. On the
basis of these findings, the diagnosis of paraneo-
plastic cerebellar degeneration was made.

The patient underwent pelvic exploration
and was found to have a right tubal carcinoma
with metastasis to the left ovary. A total abdom-
inal hysterectomy and bilateral salpingo-oopho-
rectomy was performed. Surgery was followed by
a course of chemotherapy consisting of adriamy-
cin, cisplatin, and VP16 given monthly for six
months, followed by a single treatment with car-
boplatin.

Her serum anti-Yo antibody was negative on
two occasions before treatment was begun. Carci-
noembryonic antigen was elevated to the 20-30
ng/mL range and remained elevated after ther-
apy was completed.

The patient was considered to be in complete
remission after completion of chemotherapy. A
CT scan of the abdomen and pelvis was negative.
At this time, a marked improvement in her neu-
rologic status was noted. She was able to stand
unassisted and walk with a walker. Finger-to-
nose and heel-to-shin testing returned to normal.
Rapid alternating movements were within nor-
mal limits. As of the time of publication, she re-

The Mount Sinai Journal of Medicine Vol. 60 No. 2 March 1993

163

THE MOUNT SINAI JOURNAL OF MEDICINE March 1993

164

mains clinically free of carcinoma and continues
to improve neurologically.

Discussion. Patients with PCD may have a
range of initial cerebellar signs and symptoms,
but some differences between seropositive and se-
ronegative patients seem evident. In a recent re-
view of 32 patients with PCD, the cerebellar dis-
order preceded the diagnosis of cancer in 75% of
seropositive patients and 69% of seronegative pa-
tients. In 69% of seropositive and 38% of seroneg-
ative patients, the onset of the cerebellar syn-
drome was abrupt. In the rest, it was subacute.
Vertigo and nausea followed by gait ataxia were
the most common initial symptoms in the sero-
positive patients (5).

In another recent series of 47 patients with
PCD, all seropositive patients had gait ataxia,
limb ataxia, dysarthria, and nystagmus. All sero-
negative patients had gait ataxia and the major-
ity had limb ataxia. In most seropositive patients,
the diagnosis of PCD preceded the tumor diagno-
sis and in most seronegative patients, the diagno-
sis of PCD followed the tumor diagnosis. The an-
tibody-positive patients were more likely to have
an abrupt onset (4).

PCD typically follows an unrelenting course
resulting in profound neurologic impairment. In
one recent series, 75% of patients were confined to
bed or wheelchair despite therapeutic attempts
(5). In another series, improvement was noted in
only 5 of 47 patients. Treatment of the underlying
malignancy in the 18 patients with anti-Yo anti-
body did not result in improvement in their neu-
rologic deficits. Two of these patients showed
minor improvement with plasmaphoresis. Treat-
ment of the underlying malignancy resulted in
neurologic improvement in two seronegative pa-
tients, one with non-small cell carcinoma of the
lung and one with Hodgkin's disease. A sponta-
neous remission of neurologic symptoms was
noted in another seronegative patient with Hodg-
kin's disease. Plasmaphoresis did not appear to
benefit any seronegative patient (4).

In a recent study, Furneaus et al. studied tu-
mor cells from ten seropositive patients and ten
patients with breast cancer who were normal
neurologically. They found Purkinje cell antigens
were expressed in the tumors from ten patients
with PCD but not in the tumors of the neurolog-
ically intact patients. They concluded that the
anti-Yo antibody results from an immune re-

sponse to neural antigens expressed in tumors of
patients with PCD. They further explain that
since the brain has been felt to be an immunolog-
ically privileged site, antigens expressed in neu-
ronal tissue may not have been presented to the
immune system during the establishment of im-
mune tolerance and may therefore be regarded as
"foreign" by the immune system. When these an-
tigens are expressed in tumor cells, a profound
immunologic reaction may occur (6).

This pathophysiologic mechanism strongly
suggests that early detection and treatment of the
systemic malignancy offers the greatest hope for
reversing the devastating effects of this disorder
by eliminating the inciting antigen and blunting
the ensuing immune response.

The patient presented here is unusual be-
cause her cerebellar syndrome responded so well
to treatment of the malignancy. She is an exam-
ple of early intervention leading to recovery of
neurologic function.

Acknowledgments

We thank Dr. J. B. Posner and his laboratory staff for assay-
ing anti-Yo antibody. We thank Dr. George Forster for per-
forming neurologic examination.

References

1. Croft PB, Wilkinson M. The incidence of carcinomatous

neuromyopathy in patients with various types of carci-
noma. Brain 1965; 88:427^34.

2. Cunningham J, Graus F, Anderson N, Posner JB. Partial

characterization of the Purkinje cell antigens in para-
neoplastic cerebellar degeneration. Neurology 1986; 36:
1164-1168.

3. Furneaus HM, Rosenblum M, Wong E, Woodruff P, Pos-

ner JB. Selective expression of Purkinje neuron anti-
gens in ovarian and breast tumors of patients with
paraneoplastic cerebellar degeneration. Neurology
1989; Suppl. 1:260 (abstr).

4. Anderson NE, Rosenblum MK, Posner JB. Paraneoplastic

cerebellar degeneration: clinical immunological corre-
lations. Ann Neurol 1988; 24:559-567.

5. Hammack JE, Kimmel DW, O'Neill BP, Lennon VA.

Paraneoplastic cerebellar degeneration: a clinical com-
parison of patients with and without Purkinje cell cy-
toplasmic antibodies. Mayo Clin Proc 1990; 65:1423-
1431.

6. Furneaus HM, Rosenblum M, Dalmav J, Wong E,

Woodroff P, Graus F, Posner JB. Selective expression of
Purkinje-cell antigens in tumor tissue from patients
with paraneoplastic cerebellar degeneration. N Engl J
Med 1990; 322:1844-1851.

Submitted for publication September 1992.
Final revision received December 1992.

Instructions for Authors

Communications. Address correspondence on manuscripts
review to: Managing Editor, The Mount Sinai Journal of Med-
icine, Box 1094, 50 E. 98th Street, IE, New York, NY 10029-
6574. Phone: (212) 241-6108; FAX: (212) 722-6386.

Key Words. Supply 3 to 10 key words or phrases at the bottom
of the title page, for use in indexing the article. Use Med-
ical Subject Headings from Index Medicus.

Units of Measurement. Use SI units, giving range or interval
of normal values.

Illustrations. Photographs, line drawings, graphs, and charts
should increase understanding of the text. Line drawings,
graphs, and charts should be professionally prepared. All il-
lustrations should be submitted in triplicate as sharp, high-
contrast glossy prints. Photomicrographs must be 5V8 in.
wide. On the back of each print affix a gum label with the
number of the figure (numbered consecutively in arable nu-
merals as cited in text), title of manuscript, name of senior
author, and the word "top" with an arrow indicating top edge.
Illustrations which are to appear together should be mounted
accordingly as plates and should correspond to each other in
size. Protect the prints by enclosing them in heavy cardboard;
do not use paper clips. Photographs of patients can be pub-
lished only if a copy of the permission is submitted with the
photograph. Eliminate names of patients and hospital
numbers from x-rays.

Legends. Legends should not duplicate the text. Double space.
Number each legend to match the illustration it accompanies.
Illustrations mounted as plates to appear together should be
accompanied by a single legend identifying the separate ele-
ments by position ("upper left," and so on) or by letters. If
letters are used in the legend, a corresponding letter must
appear on the print itself. Identify all abbreviations. Indicate
magnification and stain for photomicrographs.

1. Nadelman RB, Wormser GP. A clinical approach to Lyme
disease. Mt Sinai J Med 1990;57:144-156.

2. International Committee of Medical Journal Editors. Uni-
form requirements for manuscripts submitted to biomedical
journals. N Engl J Med 1991;324:424-428.

When a manuscript is accepted for publication, you will be
asked to confirm the accuracy of all references.

Reprint Orders. A reprint order form is sent to the corre-
sponding author with proofs; return it to the editorial office.

jj^Xy LISRARY^ MT.

IWfflMEDICAL CENTER

jy^^ - 3 4805 03057:?S"TJ

MOUNT SINAI JOURNAL
OF MEDICINE

Forthcoming:
Grand Rounds

Gene Therapy for Immunodeficiency and Cancer

R. Michael Blaese
National Institutes of Health

Neonatal Herpes Simplex Virus Infections:
Natural History, Pathogenesis, and Preventability

Richard J. Whitley

University of Alabama School of Medicine

Stroke Prevention In Atrial Fibrillation

Jonathan L Halperin
Mount Sinai Medical Center

Theme Issues

SEPTEMBER 1993

Quality Assurance

Editor: Suzanne Kramer

OCTOBER 1993

Cervical Disk Disease:
Controversies in Neurosurgeiy

Editors: Kalmon Post and Martin H. Savitz

The Function of the p53 Tumor Suppressor
Gene and Its Clinical Correlates
Arthur Jay Levine
Princeton University

Obscure Gastrointestinal Bleeding

Blair Lewis

Mount Sinai School of Medicine

Pharmacology of Antiinflammatoiy Agents:
A New Paradigm

Bruce N. Cronstein

New York University School of Medicine

NOVEMBER 1993

Update on Sleep Disorders

Editor: Lee K. Brown

theme section in general issue

Available now:

Toward the Conquest of Pain

Allan P. Reed, editor
volume 58. no. 3, May 1991
84 pages *15

Festschrift for Rosalyn S. Yalow:
Hormones, Metabolism, and Society

Eugene W. Straus, editor
volume 59, no. 2, March 1992
96 pages SI 5

Modem Management of
Malignant Melanoma

Michail Shafir, editor
volume 59, no. 3, May 1992
86 pages tl5

Subscriptions for 1993: $65 butloldtutls, US.; $70 all libraries; $75 Indlolduals outside US.

To order subscriptions or copies of back Issues ($15 each), please send check, payable to The Mount Sinai Journal of Medldne, to the Journal at Box 1094, Oie Guatane L Lecy
Neu) York, NY 10029-6574; counter sales at SuiU IE, 50 E. 9ath Street, New York, NY (phone: 212 241-6108; FAX: 212 722-6386).

3' THE *
OONT SINAI
F MEDiaNE

JOURNAL

).(IME 60 NCIMBER 3 MAY 1993

CONTENTS tS. S

I FRONTIERS IN GASTROENTEROLOGY ^'
In Honor of Dedication of the Dr. Henry D. Janowitz Division of 3
Gastroenterology, The Mount Sinai Hospital

David B. Sachar, Editor

Introduction David B. Sachar 165

Henry D. Janowitz and American Gastroenterology Joseph B.

Kirsner 166

Modern Concepts in Pancreatitis Peter A. Banks 170

New Frontiers In Gastrointestinal Hormones Eugene Straus 175

Current Advances in Gastrointestinal Immunology Lloyd Mayer 178

Evolution of the Controlled Clinical Trial John E. Lennard-Jones 180

Advances in Knowledge of Inflammatory Bowel Disease at Mount

Sinai Medical Center Daniel H. Present 186

Reminiscences of Henry D. Janowitz 192

GRAND ROUNDS

Obscure Gastrointestinal Bleeding Blair Lewis 200

Pharmacology of Antiinflammatory Agents: A New Paradigm

Bruce Cronstein 209

Asthma: An Inflammatory Disease Kirk Sperber 218

continued inside

Beth

Veterans Israel

Affairs Medical
ATiairs Center

The Mount Sinai Journal of Medicine is published by The Mount Sinai Medical Center of
I 11 '^^^ has the following affiliates: Beth Israel Medical Center, New York; Bronx

MOUNT SINAI
JOURNAL OF
MEDICINE

Veterans Affairs Medical Center, New York; and Elmhurst
Hospital Center, New York.

Editor-in-Chief

Sherman Kupfer, M.D.

Editor Emeritus

Lester R. Tuchman, M.D.

Associate Editors

Harriet S. Gilbert, M.D. Julius Wolf, M.D.

Managing Editor

Claire Sotnick

Business and Production Assistant

Karen Schwartz

Assistant Editors

Stephen G. Baum, M.D.
David H. Bechhofer, Ph.D.
Constanin A. Bona, M.D., Ph.D.
Edward J. Bottone, Ph.D.
Jurgen Brosius, Ph.D.
Lewis Burrows, M.D.
Joseph S. Eisenman, Ph.D.
Adrienne M. Fieckman, M.D.
Richard A. Frieden, M.D.
Steven Fruchtman, M.D.
Paul L. Gilbert, M.D.
James H. Godbold, Ph.D.

Richard S. Haber, M.D.
Noam Harpaz, M.D.
Dennis P. Healy, Ph.D.
Tomas Heimann, M.D.
Barry W. Jaffm, M.D.
Andrew S. Kaplan, D.D.S.
Samuel Kenan, M.D.
Suzanne Carter Kramer, M.Sc.
Mark G. Lebwohl, M.D.
Kenneth Lieberman, M.D.
Charles Lockwood, M.D.

Lynda R. Mandell, M.D., Ph.D.

Steven Markowitz, M.D.

Bernard Mehl, D.P.S.

Myron Miller, M.D.

Edward Raab, M.D.

Allan Reed, M.D.

Allan E. Rubenstein, M.D.

David B. Sachar, M.D.

Henry Sacks, M.D.

Robert Safirstein, M.D.

Ira Sanders, M.D.

Martin H. Savitz, M.D.
Clyde B. Schechter, M.D.
Michael Serby, M.D.
Phylhs Shaw, Ph.D.
George Silvay, M.D.
Barry D. Stimmel, M.D.
Nelson Stone, M.D.
Max Sung, M.D.
Carl Teplitz, M.D.
Rein Tideiksaar, Ph.D.
Richard P. Wedeen, M.D.

Editorial Board

Barry Freedman, M.B.A.
Richard Gorlin, M.D.
Nathan Kase, M.D.

Fanayotis G. Katsoyannis, Ph.D.
Charles K. McSherry, M.D.
Jack G. Rabinowitz, M.D.

John W. Rowe, M.D.
Alan L. Schiller. M.D.
Alan L. Silver, M.D.

Alvin S. Teirstein, M.D.
Rosalyn S. Yalow, Ph.D.

The Mount Sinai Journal of Medicine (ISSN No. 0027-2507; USPS 284-860) is published 6 times a year in January, March, May,
September, October, and November in one indexed volume by the Committee on Medical Education and Publications (Owner), The
Mount Sinai Hospital, Box 1094, One Gustave L. Levy Place, New York, NY 10029-6574. Subscription price: individuals, U.S.,
$65 per year; libraries, $70; individuals outside the U.S., $75. Single copies, $15. New subscriptions begin with the first issue of
the calendar year. Send notice of change of address 60 days before the change is effective. Second-class postage paid at New York,
NY and at additional mailing offices. POSTMASTER: Send address changes to The Mount Sinai Journal of Medicine, Box 1094, One
Gustave L. Levy Place, New York, NY 10029-6574. Copyright 1993 The Mount Sinai Hospital.

Volume 60
Number 3
May 1993

CONTENTS continued

GENERAL ARTICLES

Prevalence of Atopy in an Inner-City Asthmatic Population Kirk

Sperber, David Kendler, Lai Ming Yu, Hemal Nayak, and An-
drew Pizzimenti 227

Ollgoclonai Banding in AIDS and Hemophilia Etta B. Frankel,
Ira Z. Leiderman, Michael L. Greenberg, Sonya Makuku, and
Shaul Kochwa 232

CASE REPORTS

Gold-Induced Colitis: A Case Report and Review of the Litera-
ture Victoria Teodorescu, Joel Bauer, Simon Lichtiger, and
Mark Chapman 238

Progressive Nemaline Rod Myopathy in a Woman Coinfected
with HIV-1 and HTLV-2 Joseph Maytal, Steven Horowitz,
Stanley Upper, Bernard Poiesz, Chang Yi Wang, and Freder-
ick P. Siegal 242

ABSTRACTS

Abstracts in Medicine 247

Abstracts in Ophthalmology 250

BOOK REVIEWS

Gastric Peptide Secretion R. Cheli, A. Perasso, and G. Testino,

eds. Reviewed by J. Lawrence Werther 257

Non-Responders in Gastroenterology: Causes and Treatments

G. Dobrilla, K. D. Bardhann, and A. Steele, eds. Reviewed by

Barry W. Jaffin 257

Introduction

David B. Sachar, M.D.
The Dr. Burrill B. Crohn
Professor of Medicine
Director

Division of Gastroenterology
Mount Sinai Medical Center

Frontiers in

Gastro-

Enterology

This issue of The Mount Sinai Journal of Medicine commemorates
a historic moment in the life of the Medical Center. For the first
time in its 140-year existence, a division of Mount Sinai's Depart-
ment of Medicine has been endowed and named for an individual.
Even more meaningfully to us, this honor has been bestowed not
on a faceless donor or a shadowy figure from the distant past, but
on a colleague and contemporary, still professionally active, phys-
ically vigorous, and academically productive among us within
shouting distance of octogenarian status.

Since retiring in 1983 from his quarter-century stewardship of
Mount Sinai's Division of Gastroenterology, Henry Janowitz, clin-
ical professor emeritus and consultant physician, has continued to
play an integral role in the daily education of our gastrointestinal
fellows and to stimulate a ceaseless flow of innovative clinical
research. He is a superb clinician who has helped countless people,
a researcher who has made fundamental contributions to our un-
derstanding and treatment of gastrointestinal disease, a teacher of
some of the world's top gastroenterologists, and a friend to all of us.

In formal ceremonies on December 8, 1992, the President and
Board of Trustees of the Mount Sinai Medical Center officially
named the Dr. Henry D. Janowitz Division of Gastroenterology at
Mount Sinai. To mark the occasion, hundreds of colleagues,
present and former students, patients, and friends gathered in the
Stern Auditorium to attend an international symposium on "Fron-
tiers in Gastroenterology." Virtually the entire scientific program
of that symposium is represented in this Journal, with major sup-
port from Kabi Pharmacia, a long-standing corporate friend of ac-
ademic gastroenterology at Mount Sinai and throughout the world.

The unofficial dean of American gastroenterology, Joseph
Kirsner, leads off with a historical perspective on Henry Janowitz's
many contributions to the academic life-blood of our specialty. Spe-
cific examples follow of the seminal advances in basic and clinical
gastrointestinal pathophysiology, spearheaded by three of Dr. Jan-
owitz's former fellows: Peter Banks on pancreatitis, Eugene Straus
on gastrointestinal hormones, and Lloyd Mayer on gastrointesti-
nal immunology. Next, in a dazzlingly comprehensive yet crys-
tal-clear overview, the world's leading figure in gastrointestinal
clinical research. Professor John E. Lennard-Jones of London, dis-
cusses the evolution of the controlled clinical trial, an enterprise
that has been much enriched by the activities of Henry Janowitz
and his associates at Mount Sinai. Then, to exemplify the impor-
tant applications of clinical trials and clinical experience to daily
practice, Daniel Present concludes the formal program of the sym-
posium with a spirited review of Mount Sinai's leading role in
promoting advances in the treatment of inflammatory bowel dis-
ease.

Finally, fond reminiscences of Henry Janowitz's personal and
intellectual impact are provided by half a dozen former students
and other colleagues, from within our own walls to the farthest
reaches of North and South America, the United Kingdom, and
Australia. Indeed, five other articles in this issue — independent of
this Festschrift — deal with gastrointestinal or hepatologic topics.
This special issue of the Journal, therefore, bears witness to the
fact that Henry Janowitz's imprint on gastroenterology at Mount
Sinai — and, by extension. Mount Sinai's imprint on gastroenterol-
ogy worldwide — are indelible.

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

165

Henry D. Janowitz and
American Gastroenterology

Joseph B. Kirsner M.D., Ph.D., M.A.C.P.

I AM deeply honored to represent the gastroen-
terology community in recognizing Dr. Henry
Janowitz as physician, researcher, teacher, and
humanist, as we dedicate the Dr. Henry D. Jan-
owitz Division of Gastroenterology at the Mount
Sinai School of Medicine and The Mount Sinai
Hospital. Henry's outstanding career, now 54
years and counting, defies synopsis, and I am
privileged to review his superlative record of
achievement.

Henry David Janowitz was born in March
1915 in Paterson, NJ. His parents, with no formal
educational background, encouraged Henry and
his younger brother Morris to excel in school. The
boys grew up in a loving home, deeply imbued
with a reverence for learning and scholarship.

Henry's extraordinary intelligence and per-
severance rapidly surfaced. In high school he
wrote poetry, worked on the school paper, and in
1931, at age 16, crossed the Hudson River to take
on Columbia College. Henry studied Shakespeare
with Mark Van Doren, the poet; philosophy with
Irwin Edman (a disciple of George Santayana)
whom Henry revered; and Renaissance literature
with Raymond Weaver. He attained Phi Beta
Kappa and struggled with the opportunity to en-
ter medical school or to accept a fellowship in phi-
losophy.

Adapted with permission from J. B. Kirsner, Friedenwald
Presentation to Henry David Janowitz, M.D. Gastroenterol-
ogy 1984; 87:749-753. Adapted from the author's presentation
at the Dedication of the Dr. Henry D. Janowitz Division of
Gastroenterology on December 8, 1992 at the Mount Sinai
Medical Center. From the Department of Medicine, Univer-
sity of Chicago. Address reprint requests to the author, The
Louis Block Distinguished Service Professor of Medicine, Uni-
versity of Chicago Medical Center, MC 2100, 5841 South
Maryland Avenue, Chicago, IL 60637-1470.

Professor Edman earned himself a place, if
only a footnote, in the annals of gastroenterology
when he turned Henry away from philosophy. As
Henry remembers, "Edman said to me one day:
'In philosophy one has to be a great philosopher,
but there is always room for a good doctor.' " The
wonderful irony is that Henry became a great
physician-philosopher; a scholar who has pro-
vided inspiration and wisdom to countless col-
leagues and students. Henry's four years at the
Columbia College of Physicians and Surgeons
were a continuation of his undergraduate success.
He made Alpha Omega Alpha and in 1939 grad-
uated near the top of his class.

Henry's two years of house staff training at
Mount Sinai in New York brought him into con-
tact with many outstanding physicians. He spent
a year in pathology with Paul Klemperer; and his
imagination was fired by men like George Baehr,
Isadore Snapper, Emanuel Libman, and Bela
Schick, among others. World War II arrived and
Henry joined the Army, serving for four and one
half years. Assigned to Fort Smith, Arkansas, the
young lieutenant reported for duty on the day
that Thomas Urmy, the hospital's gastroenterol-
ogist, had been elevated to chief of medicine.
Urmy, a disciple of the late Chester Jones of Bos-
ton, immediately designated Henry as the gastro-
enterologist and himself as Henry's preceptor.

After the war, Henry knew that he needed
more training, more science, if he was to be a
"real" gastroenterologist. He correctly assessed
the trend in gastroenterology toward physiologic
research and became associated with the late
Morton Grossman in the Department of Clinical
Science at the University of Illinois in Chicago.
Henry's research produced a Master of Science
thesis on the physiology of hunger and appetite, a
field that he continues to follow. From Grossman

166

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

HDJ AND GASTROENTEROLOGY— KIRSNER

167

he also acquired some of the art of research and
the excitement and satisfaction of successful in-
vestigation.

Returning to Mount Sinai, his inquisitive
mind took him into several fruitful areas. With
the noted physiologist Franklin Hollander, he in-
vestigated electrolyte secretions of the upper gas-
trointestinal tract. With the talented surgeon
David Dreiling, he participated in highly produc-
tive research on the pancreas. Others at Mount
Sinai who influenced Henry included A. A. Berg,
the premier gastrointestinal surgeon, Richard
Marshak, the skillful gastrointestinal radiologist,
and for a short time, the brilliant Sol Berson.
Confirming the sage observation of Sir Isaac
Newton, "Discovery is never an isolated event;
each discoverer stands on the shoulders of gi-
ants," Henry throughout had the foresight and
the wisdom to associate himself with clinical and
research giants.

In 1952, while continuing his research,
Henry began his private practice in the office of
Burrill Crohn, initiating a close association that
lasted nearly 40 years. Henry opened his own of-
fice in 1958. At that time also, the late Alexander
Gutman, chief of Medicine, asked him to establish
the first division of gastroenterology at Mount
Sinai Hospital. The challenge was formidable;
but Henry's objective was clear from the start: a
program that would encourage physicians to
blend clinical investigation with excellent patient
care, on a foundation of superb teaching. Thirty-
five years later there are approximately 125 first-
class gastroenterologists throughout the country
and abroad who have been trained by Henry,
many of whom have made important contribu-
tions to clinical and academic gastroenterology.
To mention only a few: Jerome Waye (endoscopy),
Irwin Sternlieb (Wilson's disease). Jack Hansky
at Monash University in Sydney, Australia, Wal-
ter Dyck at The Scott and White Clinic in Texas,
Peter Banks of Beth Israel Hospital, Boston, Bur-
ton Korelitz of New York, Arthur Lindner of New
York University, and David Sachar, the B. B.
Crohn Professor of Medicine at Mount Sinai and
the man who succeeded Henry in July 1983 as
head of the Division of Gastroenterology at
Mount Sinai.

Research, training, care of patients, partici-
pation in local and national societies — Henry ap-
proached every task with the zeal of Tennyson's
Ulysses, determined "to strive, to seek, to find,
and not to yield." His major research interests
and accomplishments have been in hunger and
appetite, electrolyte secretion by the upper gas-
trointestinal tract, pancreatic physiology, and in-

flammatory bowel disease. He pioneered in exam-
ining the role of the upper gastrointestinal tract
in the regulation of food intake. He was the first
to document the stimulating effect of secretin on
the electrolytes of bile. He persevered in estab-
lishing the role of carbonic anhydrase in the se-
cretion of hydrogen ions by the gastric parietal
cells after this relationship had been abandoned
by prominent physiologists. His studies of gastric
function in a patient with a gastric fistula paral-
leled the classic observations of Beaumont and
others.

Dr. Janowitz was the first investigator also to
authenticate the safe inhibition of pancreatic se-
cretion by the carbonic anhydrase inhibitor Di-
amox; the first to demonstrate the direct damag-
ing effect of alcohol on the pancreas, and the first
to establish inhibition of lipoprotein lipase as one
of the mechanisms of the hyperlipidemia compli-
cating acute pancreatitis. These and other origi-
nal observations culminated in an important
book. Pancreatic Inflammatory Disease.

Henry's association with Dr. Crohn kindled a
highly productive interest in inflammatory bowel
disease, including important studies on toxic di-
latation of the colon in ulcerative colitis, statisti-
cal precision in defining recurrence rates in
Crohn's disease, the intestinal and extra-intesti-
nal manifestations of inflammatory bowel dis-
ease, the usefulness of therapeutic measures in-
cluding ACTH and hydrocortisone, and quality of
life after surgery for Crohn's disease. His approx-
imately 300 publications are models of perception
and scientific rigor.

Editing various journals constitutes yet an-
other major career of Dr. Janowitz. He has served
as editor or as a member of the editorial board of
the Handbook of the American Physiological So-
ciety, the American Journal of Chronic Diseases,
and the American Journal of Physiology, and as
chief editor of the revived American Journal of
Digestive Diseases (now known as Digestive Dis-
eases and Sciences), and the Proceedings of the
Society for Experimental Biology and Medicine.

Henry has been an active member of many
prestigious societies: the American Physiological
Society, the American Society for Clinical Inves-
tigation, the Association of American Physicians,
the Society for Experimental Biology and Medi-
cine, the American Society for Clinical Nutrition.
He has been president of the New York Gastro-
enterological Association and a member of the
American Board of Gastroenterology. He is a
founder and for seven years was the first chair-
man of the Scientific Advisory Board of the Na-
tional Foundation for Ileitis and Colitis.

168

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

He joined the American Gastroenterologic
Association (AGA) in 1952 and has been secre-
tary, vice president, and president. His tenure in
office was characterized by innovation and activ-
ity. The Second Conference on Digestive Diseases
as a National Problem (April 23-25 and May 18,
1973) was initiated by Dr. Janowitz, "who be-
lieved that the field of digestive disease should be
re-examined and perhaps stimulated as it was by
the earlier conference of 1967."

In 1973 also, Dr. Janowitz created the AGA
Presidential Commission on the Status and Fu-
ture of the AGA, with the charge "to undertake a
critical review of the Association's organization,
functions, and policies, financial resources, the
relationship of the AGA with the National Insti-
tute of Arthritis, Metabolism, and Digestive Dis-
eases, the government, and with international
gastroenterology, the possibility of a National
Voluntary Organization for Digestive Diseases,
and future goals. The Janowitz Presidential Com-
mission is credited with stimulating the recent
spectacular growth of the AGA, and its recom-
mendations continue to influence gastroenterol-
ogy today.

Involvement with the AGA has been one of
the great pleasures of Henry's scientific and per-
sonal life: "It was not merely the exchange of in-
formation or even doing good for gastroenterol-
ogy, but the personal contacts with my many
friends over the years that provided such a rich
source of satisfaction." In 1984 I had the distinct
pleasure, on behalf of the American Gastroenter-
ological Association, to present to Henry its most
treasured honor, the Julius Friedenwald Medal,
in recognition of his many contributions to the
science and practice of gastroenterology.

Henry is in demand as a speaker in this coun-
try and abroad. He has been the McArthur Lec-
turer at the University of Edinburgh, the Comfort
Lecturer at the Mayo Foundation, and visiting
professor at many medical centers and universi-
ties.

A major strength of Henry's career has been
the great supportive and complementary role
played by his accomplished family, his wife Ade-
line (Addie), and their two daughters, Mary and
Annie. His late brother, Morris, one of the na-
tion's leading sociologists, was the Lawrence A.
Kimpton Distinguished Service Professor at The
University of Chicago.

Addie Janowitz is a student of trends in mod-
ern art and is a widely recognized Henry James
scholar. She is the editor of the Henry James
Journal and the Journal of Pre-Raphaelite Stud-
ies. Throughout, Addie has been a colleague and

champion of Henry's work and her literary activ-
ities have maintained Henry's scholarly interest
in literature.

Their daughter Mary, a graduate of Barnard
College, lives in San Francisco with her husband
and their two children. Annie studied at Oxford
and has a Ph.D. in English literature from Stan-
ford University. She has taught at Smith College
and is a newly appointed assistant professor at
Brandeis University. These family activities pro-
vide Henry access to the latest in literary
thought. His intellectual curiosity is enormous;
and a satisfying vacation consists of settling down
with Fitzgerald, Hemingway, or Faulkner.

Henry's impressive bibliophilic talents would
easily qualify him as the literary critic of The
Times of London, the Manchester Guardian, or
the New York Times. Indeed, such contributions
as "Marius the Epicurean in T. S. Eliot's Poetry,"
"An Anglo-American Consultation: Sir William
Osier Refers Henry James to Sir James MacKen-
zie," (with Adeline R. Tintner), and "Falstaffs
Nose Was 'A Table of Green Fields': A Footnote to
Ephim Fogel's Defense of the Folio Reading in
Cahiers Elisabethains" place Janowitz in the
company of such distinguished physician-littera-
teurs as Oliver Goldsmith, A. Conan Doyle, and
W. Somerset Maugham. And they lacked his med-
ical and scientific achievements.

This brief account only hints at Henry's
achievements. Like Henry James, his artistry has
been subtle and deep. In moments of self-evalua-
tion, he claims to be able to "only think about one
thing at one time," which for most of us would be
a considerable feat. He obviously has thought
about many things, each in their own time; in
each instance the outcome has been new knowl-
edge, and a brightening of the path of scientific
endeavor.

Having turned over the leadership of his di-
vision at Mount Sinai School of Medicine to David
Sachar, he remains as clinical professor of medi-
cine, co-director of the Gastroenterology Disease
Unit, and director of the Burrill Crohn Founda-
tion. He continues his "normal" 12-hour daily
schedule as a clinician, consultant, teacher, clin-
ical investigator, and author.

His sought-after clinical preceptorship, ac-
commodating four fellows annually, each for
three months, is a model of personalized training.
The fellow begins the day at 7:30 am in Dr. J's
office, then accompanies Henry on hospital visits,
rounds, and conferences. Henry's 1992 paper,
"The Art of the Gastrointestinal Consultation,"
reflects this extensive clinical experience, and
his conclusion that "gastroenterology is still an

Vol. 60 No. 3

HDJ AND GASTROENTEROLOGY— KIRSNER

169

active transaction between doctor and patient"
epitomizes the ethos of gastroenterology. Henry's
clinical research program currently emphasizes
the metanalysis of clinical data on the natural
history and the treatment of IBD. His 1985 pub-
lication, Inflammatory Bowel Disease: A Personal
View, reflects this long experience. His clinical
achievements have been recognized in the 1986
Stuart Distinguished Lecture and in the 1992
Clinical Achievement Award from the American
College of Gastroenterology. His contributions to
the medical literature continue: Your Gut Feel-
ings: A Complete Guide to Living Better with In-
testinal Problems (1987), Indigestion (1992), and
another on IBD scheduled for 1993 publication.

How does he do it? Where does he get the
motivation and the energy to accomplish so
much? A partial answer probably is to be found in
what William Osier considered a master-word in
medicine: Work, "the open sesame to every portal;
the true philosopher's stone which transmutes
the base metal of humanity into gold . . . and the

touchstone of progress." Another part of the
answer is in Henry Janowitz's unwavering com-
mitment to excellence, searching imagination,
disciplined inquiry, reasoned judgment, and ap-
preciation of originality.

My own admiration for Henry is unbounded.
He has demonstrated over the years the qualities
of intellect, commitment, and integrity that be-
speak the philosopher-physician, a valued tradi-
tion. He has shown that important research could
be carried out successfully by physicians profi-
cient in both the art and the science of patient
care, and that this combination enhances the clin-
ical relevance of the research.

Henry, the dedication of the Division of Gas-
troenterology in your name is a well-deserved and
uniquely appropriate recognition of your out-
standing accomplishments, and also honors the
Mount Sinai School of Medicine. The entire gas-
troenterologic community joins me in extending
to you our warmest congratulations and best
wishes for the future.

Modern Concepts in Pancreatitis

Peter A. Banks, M.D.

Although I completed my fellowship in gastro-
enterology at The Mount Sinai Hospital 25 years
ago, I never really left Mount Sinai. On a per-
sonal level, friendships I developed while in train-
ing have been an important part of my life for
the past 25 years, and I trust will continue for at
least another 25 years. On a professional level,
the lessons I learned at Mount Sinai will endure
for a lifetime. The educational environment was
stimulating and exciting, and it included input
not only by gastroenterologists, but also by sur-
geons, radiologists, pathologists, nutritionists,
and many other specialists.

I would call attention to two lessons in par-
ticular that are the special legacy of Dr. Janowitz.
The first lesson was to think critically. Whether
presenting at our weekly physiology seminar on
Saturday morning or at the bedside, you had to
know your stuff, and you had to think critically.
The second lesson was to develop intellectual cu-
riosity. Largely due to the impact that Dr. Jan-
owitz had on all of his fellows, we learned to probe
deeper, learn more, and do more. These lessons,
and many more taught by Dr. Janowitz, inspired
me to make the additional efforts to teach and to
engage in clinical research. It was at The Mount
Sinai Hospital that I was stimulated to explore
the mysteries of the pancreas.

The field of pancreatitis has advanced rap-
idly. In acute pancreatitis, new information has
been gathered on the cell biology of pancreatitis
(1, 2) and mediators of inflammation (3-16).
There have also been advances in the diagnosis of
acute pancreatitis by CT scan (17-20), early rec-

Adapted from the author's presentation at the Dedication of
the Dr. Henry D. Janowitz Division of Gastroenterology on
December 8, 1992 at the Mount Sinai Medical Center. From
the Clinical Gastroenterology Service, Brigham and Women's
Hospital, 75 Francis Street, Boston, MA 02115. Address re-
print requests to the author, Director of the Clinical Gastro-
enterology Service, there.

ognition of infected necrosis by guided percutane-
ous aspiration (21-23), increased awareness of
the threat of sterile necrosis (24-27), and im-
provements in treatment (21, 24, 28-32). In the
field of chronic pancreatitis, there is increased
understanding of the causes of pain (33-38) and
potential improvements in the treatment of pain
(39-43). These advances will be described briefly.

Acute Pancreatitis

Cell Biology. Digestive enzymes as well as
other proteins, including lysosomal hydrolases,
are synthesized on ribosomes attached to the
rough endoplasmic reticulum and transported to
the Golgi complex. As they traverse the Golgi
complex, hydrolases that are transported to lyso-
somes are sorted from other proteins via chemical
alterations and are then bound to mannose-6-
phosphate receptors. Once they reach the lyso-
somal compartment, the lysosomal hydrolases
dissociate from their receptors. In comparison, di-
gestive enzymes pass through the Golgi complex,
are packaged in condensing vacuoles, mature into
zymogen granules, and migrate to the apical cell
surface. Zymogens then fuse to the apical cell sur-
face and discharge enzymes into the lumen.

A number of protective mechanisms prevent
premature activation of pancreatic enzymes.
First, maturation of digestive enzymes and lyso-
somal hydrolases occur via separate pathways. If
they travelled along the same pathway, the lyso-
somal enzyme, cathepsin-p, could potentially ac-
tivate trypsinogen to trypsin. Second, digestive
enzymes and lysosomal hydrolases travel within
membrane-enclosed compartments. Third, diges-
tive enzymes that have the potential of injuring
acinar cells are synthesized, transported, and se-
creted as proenzymes. Fourth, a trypsin inhibitor
travels with digestive enzymes to prevent the pre-
mature activation of trypsinogen to trypsin.
Fifth, pH within zymogen granules is acidic. An

170

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

PANCREATITIS— BANKS

171

acid pH inactivates any trypsin that might be
prematurely activated from trypsinogen. Sixth,
within the lumen of the pancreatic ducts, any
trypsin that might be activated is either inhibited
by the trypsin inhibitor or degraded in the alka-
line pH.

Several experimental models of acute pancre-
atitis produce similar injury within acinar cells
(1, 2). In one model, that of diet-induced pancre-
atitis, mice fed a diet that is choline-deficient and
ethionine-enriched develop fatal pancreatic ne-
crosis. In another model, that of secretagogue-in-
duced pancreatitis, rats administered a supra-
maximal dosage of the cholecystokinin analogue,
cerulein, develop a reversible form of interstitial
edematous pancreatitis.

In both of these models, the pathophysiology
of acinar cell injury takes a similar form. The
initial event leading to this injury is a blockage of
secretion resulting in accumulation of zymogen
granules within acinar cells. The next step is a
fusion of lysosomes and zymogens within large
vacuoles. Activation of pancreatic enzymes pre-
sumably then takes place because these vacuoles
now contain both lysosomal hydrolases (including
cathepsin-p) and digestive enzymes. Once cathep-
sin-3 activates trypsinogen to trypsin, trypsin can
then catalyze many proenzymes to their active
form. Since these vacuoles that contain both ly-
sosomal hydrolases and pancreatic enzymes are
quite fragile, activated enzymes are then released
into the cytoplasm and cause intracellular injury.

Colocalization of lysosomal and digestive en-
zymes can also be demonstrated in acinar cells of
the rabbit or rat following pancreatic ductal liga-
tion, but acute inflammation does not occur.
Hence, there appears to be a spectrum of inflam-
matory responses when there is colocalization,
ranging from absent (no inflammation) to mild
interstitial edema to fatal necrotizing pancreatitis.

The relevance of these findings to human
pancreatitis remains uncertain. Additional ex-
perimentation will be needed to determine
whether these experimental models accurately
reflect the earliest changes in human pancreati-
tis, and if they do, whether a strategy can be de-
veloped to prevent intracellular injury.

Mediators of Inflammation. It is known that
pancreatic enzymes such as phospholipase-Aa and
elastase injure pancreatic parenchyma. A variety
of synthetic inhibitors of trypsin, inhibitors of
other proteases, and inhibitors of phospholipase-
Ag have been shown in experimental pancreatitis
to reduce inflammation and at times improve sur-
vival, but studies in humans have not yet been
conducted (3-6).

Also, a variety of mediators of inflammation
may play a role in intensifying or perpetuating an
inflammatory response:

• Oxygen-derived free radicals

• Platelet-activating factor

• Neutrophils

• Leukotrienes

• Complement

• Cytokines

• Prostaglandins

• Endotoxin

The role of these potential mediators in acute
pancreatitis remains to be clarified. For example,
in some reports on experimental pancreatitis but
not others, oxygen-derived free radicals have
been thought to be significant mediators of cell
injury (7-9). In addition, inhibitors of neutrophil
migration and antagonists of platelet-activating
factor have been shown experimentally to reduce
inflammatory response (10, 11). Activation of
complement may play a role in the development
of multiple system organ failure (12). In addition,
there has been speculation that acute pancreati-
tis may induce a cytokine response that intensi-
fies the inflammatory response and increases sus-
ceptibility to sepsis. In one report, patients with
severe acute pancreatitis exhibited very high lev-
els of interleukin-6 in serum (13). Studies with
prostaglandins have yielded different results on
the natural history of experimental pancreatitis.
Some prostaglandins appear to be helpful,
whereas others increase microvascular perme-
ability and increase the severity of pancreatitis
(14-16). Thromboxane- A2, by its properties as a
vasoconstrictor and stimulator of platelet aggre-
gation, appears to worsen experimental pancre-
atitis, and its effects can be blocked by a potent
inhibitor (14).

In summary, a considerable amount of work
will be required to define the role of these poten-
tial mediators of inflammation in the natural his-
tory of acute pancreatitis and to develop ways to
counteract their impact.

Role of CT Scan. The use of CT scan with
intravenous contrast utilizing bolus technique
has had an enormous impact on our ability to care
for patients with acute pancreatitis (17-19). Fol-
lowing intravenous contrast with bolus tech-
nique, uniform perfusion of the pancreas indi-
cates that the microcirculation of the pancreas is
intact and that the process is interstitial pancre-
atitis. When portions of the pancreas enhance
poorly following bolus injection, these poorly per-
fused or nonperfused areas are thought to repre-
sent areas of necrosis or fluid. The larger the ar-

172

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

eas of nonperfusion, the more confident is the
clinician that the patient has necrotizing pancre-
atitis. It is much more difficult to ascertain on CT
scan whether inflammatory changes in the peri-
pancreatic tissue represent fluid or necrotic fat. If
Hounsfield units are in the range of 0-15, the
likelihood is that these areas represent fluid.
When Hounsfield units exceed 15, peripancreatic
necrosis is more likely.

The significance of pancreatic necrosis lies in
the increased risk of mortality associated with it
(20). In interstitial pancreatitis, even when asso-
ciated with systemic complications, mortality
should be less than 5%, assuming proper care of
the patient. Mortality in necrotizing pancreatitis
is much higher. The overall mortality in sterile
necrosis is 10%, but in infected necrosis it is 30%.
Mortality in necrotizing pancreatitis increases
further in the presence of systemic complications.
That is, in sterile necrosis with severe systemic
complications such as shock, respiratory insuffi-
ciency, or renal insufficiency, the mortality may
be 30%-40%. In infected necrosis, with similar
systemic complications, the mortality may be
greater than 50%. Hence, the interplay between
CT appearance and the presence or absence of
systemic complications helps the clinician assess
the severity of acute pancreatitis.

Infected Necrosis. Many if not most deaths
in patients who have acute pancreatitis are asso-
ciated with infected necrosis of the pancreas (20).
One important reason for the high mortality is
delay in surgical debridement. During the past
ten years, Stephen Gerzof and I have worked to-
gether establishing the role of guided percutane-
ous aspiration in the early diagnosis of infected
necrosis. In our updated experience with 87 pa-
tients strongly suspected of harboring pancreatic
infection on the basis of a CT scan showing severe
pancreatitis (Balthazar-Ranson score of D or E),
extreme systemic toxicity (usually a very high
white count of 20,000-30,000 and temperature
102°F-104°F), 44/87 patients were indeed in-
fected.

We have learned that guided percutaneous
aspiration is a safe accurate method in the early
diagnosis of pancreatic infection, that infection
occurs early in acute pancreatitis, that both fluid
and necrotic tissue are infected, and that early
surgical debridement offers the best chance in re-
ducing mortality (21-23).

Sterile Necrosis. As I have continued to eval-
uate the outcome of patients with necrotizing
pancreatitis, I have been impressed that patients
with severe sterile necrosis who appeared ex-

tremely toxic on the basis of systemic complica-
tions also have high mortality. I have sought to
identify prognostic factors that might help ex-
plain this high mortality. As a result of observa-
tions on 26 patients with sterile necrosis, all of
whom had at least one systemic complication, I
have established several unfavorable prognostic
factors that correlate with high mortality (24),
including:

• Very high Ranson's score and APACHE-II
scores within the first 48 hours

• Development of multiple systemic complica-
tions

• In particular, the development of shock, which
was the best predictor of a fatal outcome; in our
experience, almost all patients who eventually
died experienced shock within the first 7-10
days of hospitalization.

On the basis of this information, we can now
sort out patients who can be expected to survive
with intensive care from those who have a high
likelihood of a fatal outcome. Although the role of
surgery in sterile necrosis of the pancreas re-
mains controversial, in my view, patients with
unfavorable prognostic factors should undergo
surgical debridement until such time as more ef-
fective medical treatment is found that improves
mortality.

In the context of sterile pancreatic necrosis, it
has become increasingly clear that a fluid-density
mass may replace pancreatic parenchyma and be
erroneously interpreted as pancreatic pseudocyst
(25, 26). In our experience with 17 patients, al-
most all intrapancreatic fluid-density masses con-
tain a considerable amount of necrotic debris as
well as fluid (27).

The therapeutic implication of this finding is
important. Whereas pancreatic pseudocysts that
extend out of the pancreas contain mostly fluid
and only a scant amount of necrotic debris and
can be treated by surgical decompression and at
times by either pigtail catheter drainage or endo-
scopic cyst-gastrostomy, neither endoscopic nor
radiologic techniques should be attempted for in-
trapancreatic fluid density masses. The reason is
that these techniques are not able to remove ne-
crotic debris. Failure to eliminate this debris
leads to secondary intrapancreatic infection. In
our experience, all patients with intrapancreatic
fluid density masses treated by surgical debride-
ment did well; those treated with endoscopic or
radiologic techniques became secondarily in-
fected and eventually required surgical debride-
ment.

Vol. 60 No. 3

PANCREATITIS— BANKS

173

Treatment. Infected necrosis requires surgi-
cal debridement. In sterile necrosis, the role of
surgical debridement remains controversial (24).
Prophylactic antibiotics have recently been
shown to reduce the incidence of pancreatic infec-
tion but not to improve mortality (28). Endoscopic
sphincterotomy has been recommended in severe
gallstone pancreatitis (29). Although morbidity
was improved in elderly patients with severe gall-
stone pancreatitis, it is possible that this im-
provement related more to eradication of biliary
sepsis than to improvement in the pancreatitis
itself. Efforts should be made to protect the mi-
crocirculation of the pancreas (30, 31). In one
study, low-dose dopamine was helpful in the ex-
perimental animal in reducing microvascular
permeability (30). Methods to improve phagocytic
function may prove to be helpful (32).

Chronic Pancreatitis

Causes of pain in chronic pancreatitis include
superimposed acute parenchymal inflammation,
neural inflammation, and increased intrapancre-
atic pressure (33-38). It appears that pancreatic
nerves are not trapped or squeezed by fibrosis but
are actually larger than normal (34). In addition,
elegant ultrastructure studies have shown that
there is a disruption of the perineurial sheath
that shields the internal neural components from
surrounding tissue (34). Once this sheath has
been disrupted, inflammatory cells can then in-
vade nerve tissue and may evoke an inflamma-
tory response by discharging enzymes directly
into nerve bundles.

In addition, there is increasing evidence that
intrapancreatic pressure is increased in acute
pancreatitis (36-38). This increase may be suffi-
cient to reduce capillary filling and even induce
tissue acidosis (38). It is conceivable that pain re-
sults from these phenomena and may possibly in-
volve the formation of oxygen-derived free radi-
cals (39).

Treatment of Pain. The treatment of pain in
chronic pancreatitis includes discontinuation of
alcohol and the administration of analgesics.

If the pain has as its basis an ischemic event,
allopurinol may play a role in its treatment (39).
We are now embarking on a randomized, prospec-
tive, double-blind trial of the use of allopurinol in
treating the pain of chronic pancreatitis.

A third method is to suppress secretion with
drugs (omeprazole or an blocking agent) or
pancreatic enzymes. Although some reports have
reported encouraging results with pancreatic en-

zymes, their effectiveness seems best among pa-
tients with idiopathic pancreatitis and much less
among those with alcoholic pancreatitis (33).

Efforts to modify neural transmission by
drugs or a nerve block have, in general, not been
effective.

Methods to relieve obstruction now include
the use of stents (40, 41), lithotripsy (42), and sur-
gery. The use of stents to overcome a stricture has
become widespread. There is increasing concern
that stents may induce ductal changes that re-
semble those of chronic pancreatitis (43). Litho-
tripsy has also been reported to be beneficial in
the treatment of pain of chronic pancreatitis (42).
Although these studies are encouraging, they
have not been performed in a randomized, pro-
spective, double-blind fashion. Hence, among the
various proposed therapies to overcome ductal ob-
struction, none has been convincingly shown to be
effective.

Summary

A great deal remains to be learned about the
pathophysiology, natural history, and treatment
of acute and chronic pancreatitis. As we consider
problems in treatment, we would be well advised
to consider these words of Dr. Henry Janowitz:
"Gastroenterologists are physicians who know
how to think about and manage patients who
have gastrointestinal problems." (44). To the ex-
tent that we heed these words and always give
serious thought to what we are trying to achieve,
we will provide greater help for our patients.

References

1. Steer ML. How and where does acute pancreatitis begin?

Arch Surg 1992; 127:1350-1353.

2. Bettinger JR, Grendell JH. Intracellular events in the

pathogenesis of acute pancreatitis. Pancreas 1991;
6:S2-S6.

3. Suzuki M, Isaji S, Stanten R, Frey CF, Ruebner B. Effect

of protease inhibitor FUT-175 on acute hemorrhagic
pancreatitis in mice. Int J Pancreatol 1992; 11:59-65.

4. Lankisch PG, Pohl U, Goke B, et al. Effect of FOY-305

(camostate) on severe acute pancreatitis in two experi-
mental animal models. Gastroenterology 1989; 96:193-
199.

5. Wisner JR Jr, Renner IG, Grendell JH, Niederau C, Fer-

rell LC. Gabexate mesilate (FOY) protects against cer-
uletide-induced acute pancreatitis in the rat. Pancreas
1987; 2:181-186.

6. Tani S, Otsuki M, Itoh H, et al. The protective effect of the

trypsin inhibitor urinastatin on cerulein-induced acute
pancreatitis in rats. Pancreas 1988; 3:471^76.

7. Rutledge PL, Saluja AK, Powers RE, Steer ML. Role of

oxygen-derived free radicals in diet-induced hemor-
rhagic pancreatitis in mice. Gastroenterology 1987; 93:
41-47.

174

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

8. Schoenberg MH, Buchler M, Beger HG. The role of oxygen

radicals in experimental acute pancreatitis. Free Rad-
ical Biol Med 1992; 12:515-522.

9. Cassone E, Maneschi EMT, Faccas JG. Effects of allopu-

rinol on ischemic experimental pancreatitis. Int J Pan-
creatol 1991; 8:227-134.

10. Ais G, Lopez-Farre A, Gomez-Garre DN, et al. Role of

platelet-activating factor in hemodynamic derange-
ments in an acute rodent pancreatic model. Gastroen-
terology 1992; 102:181-187.

11. Dabrowski A, Gabryelewicz A, Chyczewski L. The effect of

platelet activating factor antagonist (BN 52021) on ce-
rulein-induced acute pancreatitis with reference to ox-
ygen radicals. Int J Pancreatol 1991; 8:1-11.

12. Roxvall LI, Bengtson A, Heideman JMI. Anaphylatoxins

and terminal complement complexes in pancreatitis.
Arch Surg 1990; 125:918-921.

13. Leser HG, Gross V, Scheibenbogen C, Heinisch A. Eleva-

tion of serum interleukin-6 concentration precedes
acute-phase response and reflects severity in acute pan-
creatitis. Gastroenterology 1991; 101:782-785.

14. Van Ooijen B, Kort WJ, Tinga CJ, Wilson JHP. Signifi-

cance of thromboxane A2 and prostaglandin I2 in acute
necrotizing pancreatitis in rats. Dig Dis Sci 1990; 35:
1078-1084.

15. Buscail L, Bussenot I, Bouisson M, et al. Protective effect

of misoprostol, a synthetic prostaglandin analog, on
cerulein-induced acute pancreatitis in rats. Pancreas
1990; 5:171-176.

16. Harvey MH, Wedgwood KR, Reber HA. Vasoactive drugs,

microvascular permeability, and hemorrhagic pancre-
atitis in cats. Gastroenterology 1987; 93:1296-1300.

17. Balthazar EJ, Robinson DL, Megibow AJ, Ranson JHC.

Acute pancreatitis: value of CT in establishing progno-
sis. Radiology 1990; 174:331-336.

18. Bradley EL III, Murphy F, Ferguson C. Prediction of pan-

creatic necrosis by dynamic pancreatography. Ann
Surg 1989; 210:495-504.

19. Larvin M, Chalmers HE, McMahon MJ. Dynamic con-

trast-enhanced computed tomography: a precise tech-
nique for identifying a localizing pancreatic necrosis.
Br Med J 1990; 300:1425-1428.

20. Banks PA. Predictors of severity in acute pancreatitis.

Pancreas 1991; S7-S12.

21. Gerzof SG, Banks PA, Bobbins AH, et al. Early diagnosis

of pancreatic infection by CT guided aspiration. Gastro-
enterology 1987; 93:1315-20.

22. Banks PA, Gerzof SG, Chong FK, et al. Bacteriologic sta-

tus of necrotic tissue in necrotizing pancreatitis. Pan-
creas 1990; 5:330-333.

23. Banks PA. Infected necrosis: morbidity and therapeutic

consequences. Hepato-Gastroenterol 1991; 38:116-119.

24. Karimgani I, Porter KA, Langevin RE, Banks PA. Prog-

nostic factors in sterile pancreatic necrosis. Gastroen-
terology 1992; 103:1636-1640.

25. Banks PA, Gerzof SG, Sullivan JG. Central cavitary ne-

crosis: differentiation from pancreatic pseudocysts on
CT scan. Pancreas 19088; 3:83-88.

26. Casey JE, Porter KA, Langevin RE, Banks PA. Clinical

features and natural history of central cavitary necro-
sis. Pancreas 1993; 8:141-145.

27. Hariri M, Slivka A, Carr-Locke DL, Banks PA. Do intra-

pancreatic fluid-density masses contain fluid or ne-
crotic tissue? Gastroenterol 1993 (Abstr) in press.

28. Bassi C, Vesentini S, Nifosi F, et al. Pancreatic abscess

and other pus-harboring collections related to pancre-
atitis: a review of 108 cases. World J Surg 1990; 14:
505-512.

29. Neoptolomous JP, Carr-Locke DL, London NJ, et al. Con-

trolled trial of urgent endoscopic retrograde cholangio-
pancreatography and endoscopic sphincterotomy versus
conservative treatment for acute pancreatitis due to
gallstones. Lancet 1983; 2:979-983.

30. Karanjia ND, Widdison AL, Lutrin FJ, et al. The antiin-

flammatory effect of dopamine in alcoholic hemorrhagic
pancreatitis in cats. Gastroenterology 1991; 101:1635-
1641.

31. Bockman DE. Microvasculature of the pancreas. Int J

Pancreatol 1992; 12:11-21.

32. Widdison AL, Karanjia ND, Alvarez C, Reber HA. Influ-

ence of levamisole on pancreatic infection in acute pan-
creatitis. Am J Surg 1992; 163:100-104.

33. Banks PA. Management of pancreatic pain. Pancreas

1991:6(Suppl 1):S52-S59.

34. Bockman DE, Buchler M, Malfertheiner P, Beger HG.

Analysis of nerves in chronic pancreatitis. Gastroenter-
ology 1988; 94:1459-1469.

35. Buchler M, Weihe E, Friess H, et al. Changes in peptider-

gic innervation in chronic pancreatitis. Pancreas 1992;
7:183-192.

36. Ebbehoj N, Borley L, Madsen P, Matzen P. Pancreatic

tissue fluid pressure during drainage operations for
chronic pancreatitis. Scand J Gastroenterol 1990; 25:
1041-1045.

37. Jalleh RP, Aslam M, Williamson RCN. Pancreatic tissue

and ductal pressures in chronic pancreatitis. Br J Surg
1991; 78:1235-1237.

38. Reber HA, Karanjia ND, Alvarez C, et al. Pancreatic

blood flow in cats with chronic pancreatitis. Gastroen-
terol 1992; 103:652-659.

39. Salim AS. Role of oxygen-derived free radical scavengers

in the treatment of recurrent pain produced by chronic
pancreatitis. Arch Surg 1991; 126:1109-1114.

40. Huibregtse K, Schneider B, Vrij AA, Tytgat GNJ. Endo-

scopic pancreatic drainage in chronic pancreatitis. Gas-
trointest Endoscopy 1988; 34:9-15.

41. Sherman S, Lehman GA, Hawes RH, et al. Pancreatic

ductal stones: frequency of successful endoscopic re-
moval and improvement in symptoms. Gastrointest En-
doscopy 1991; 37:511-517.

42. Delhaye M, Vandermeeren A, Baize M, Cremer M. Extra-

corporeal shock-wave lithotripsy of pancreatic calculi.
Gastroenterol 1992; 102:610-620.

43. Kozarek RA. Pancreatic stents can induce ductal changes

consistent with chronic pancreatitis. Gastrointest En-
doscopy 1990; 36:93-95.

44. Janowitz HA. The art of the gastrointestinal consultation:

some aphorisms. Mt Sinai J Med 1992; 59:341-349.

New Frontiers in
Gastrointestinal Hormones

Eugene Straus, M.D.

History is alive at Mount Sinai Medical Center,
and even those of us who have played the smallest
roles take interest and pride in that process. Gas-
troenterologists can take special pride, as this
event and all that it signifies calls attention to
rich traditions in the area of digestive diseases.
As one who trained at Mount Sinai, and I am
honored to be a representative of the many out-
standing gastroenterologists who drank at this
well of knowledge, curiosity, and fellowship, deep
feelings are aroused when I look around this room
and see my teachers, people who are still inform-
ing and inspiring us. Henry Janowitz is foremost
in this tradition.

Dr. Kirsner speaks of Henry's contributions
to gastroenterology. We each know many of these,
and others, because he made so many important
contributions, and I am thinking now of his early
work in gastric secretion of mucus, which I re-
cently came across in Horace Davenport's forth-
coming History of Gastric Secretion and Diges-
tion.

We also each cherish our personal experi-
ences with this man of extraordinary experience
and wisdom. I remember my first meeting with
Dr. Janowitz. In my naivete I was expecting to
discuss inflammatory bowel disease in a setting of
opulence and refined taste. I was pleasantly sur-
prised to find him working on a manuscript in a
most dingy little cubby on what I remember to be
the third floor of Atran. We discussed "Washing-
ton Square" and The Real Thing. There was an

Adapted from the author's presentation at the Dedication of
the Dr. Henry D. Janowitz Division of Gastroenterology on
December 8, 1992 at the Mount Sinai Medical Center.

From the SUNY-Health Science Center Brooklyn, N.Y.
Address correspondence and reprint requests to the author,
who is Professor of Medicine and Chief, Digestive Diseases,
Box 1196, 450 Clarkson Avenue, Brooklyn, NY 11203-2098.

abundance of refined taste, and it was all the pro-
fessor's.

On such an occasion it is well to try to put
things into perspective. Let me review briefly the
march to New Frontiers in gastrointestinal hor-
mones before trying, even more briefly and cer-
tainly with even less authority, to look down the
road ahead.

The Old Frontier

It was the first piece of good Yankee science
and its symbolic beginning came when the farm-
er's son from Lebanon, Connecticut declined an
offer of land to farm for himself and rode out on
the new mare, fondling the $100 in the leather
pouch which was the second part of papa's parting
gift. William Beaumont's formal training was a
two-year apprenticeship with Dr. Benjamin
Chandler of St. Albans, Vermont, beginning in
1810. His independent and adventurous nature,
along with his dedication to his patient, the re-
luctant Alexis St. Martin, and to meticulous ob-
servation, begin and define the great tradition of
physician investigators in this country. His Ex-
periments and Observations on the Gastric Juice
and the Physiology of Digestion, published by
F. P. Allen of Plattsburg, NY, in 1833, was read
throughout the world and opened the modern
study of digestive processes. Beaumont stimu-
lated countless physician investigators, including
Osier and Henry's professor, Andrew Conway Ivy,
both of whom wrote introductions to Myer's Life
and Letters of Dr. William Beaumont. I read the
Experiments and Observations as a first-year fel-
low and remember discussing it with Dr. Janow-
itz, whom I revered in the direct chain from Beau-
mont through my own professors: Janowitz,
Schaffner, Sachar, Lieber, Levine, Berson, and
Yalow, and I was inspired to follow. If our country

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

175

176

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

cannot restore a greater sense of purpose and sta-
bility to life within this tradition, there will be
fewer New Frontiers here.

Pavlov's extensive studies of digestion cen-
tered around the issue of regulation of functions,
and by 1902 his doctrine of "nervism" was widely
accepted. At University College in London
William Bayliss and Ernest Starling decided to
denervate the dog pancreas and, on the "great
afternoon" of 16 January 1902, the brothers-in-
law achieved regulation of water and bicarbonate
secretion by injecting extracts of duodenal mu-
cosa. Starling declared that they had discovered
"a chemical reflex." They began describing their
findings before the Royal Society within a week
(1, 2). They called their chemical messenger se-
cretin, and some three years later, at the sugges-
tion of William Hardy, Starling coined the term
"hormone."

Pavlov, always the scientist, had someone in
his laboratory repeat the secretin experiment and
quickly concluded that Bayliss and Starling were
right. While most were convinced that the duode-
num released a chemical messenger in response
to acid, there were skeptics, and even Ivy consid-
ered that secretagogues might be contained
within food until his transplantation experiments
of 1927 (3).

Standing on the shoulders of giants, John
Sydney Edkins of St. Bartholomew's Hospital
Medical School set out to find a "gastric secretin,"
and by 1905 he reported the discovery of what he
called "gastrin" to the Royal Society (4). We know
too that by 1920 there was general agreement
that gastrin did not exist, that Edkins' crude ex-
tracts contained histamine, and that it was hista-
mine which had caused his cats to secrete acid.
Poor Edkins, he had done good work, he had noted
the vasodilator activity of both his oxyntic and
antral mucosal extracts, and the fact that only
the antral extracts stimulated secretion. Cer-
tainly both extracts had contained histamine, but
intravenous injection of histamine in the concen-
trations likely to have been present in his ex-
tracts does not stimulate acid secretion. I believe
that he died in a state of scientific rejection. Post-
mortem rehabilitates.

In 1928 the cross-circulation experiments of
Ivy and Oldberg extended the earlier observation
of Okada who, working in Starling's laboratory,
had found that intestinal acidification resulted in
contraction of the gallbladder (5, 6). Active mate-
rial was extracted from duodenal mucosa and was
named cholecystokinin (CCK). Fifteen years later
Harper and Raper demonstrated that intravenous
injection of intestinal extracts stimulated the se-

cretion of pancreatic enzymes (7). They named
the hormonal material responsible for this activ-
ity "pancreozymin" (PZ). In 1966 Jorpes and Mutt
purified the crude material and demonstrated
that CCK and PZ were the same peptide (8). It is
now conventional to use the term CCK.

The Middle Frontier

One might consider that the Old Frontier in
the field of gastrointestinal hormones began with
the discovery of secretin and the conceptualiza-
tion of the endocrine system by Bayliss and Star-
ling and extended through the appreciation of the
three well-accepted gastrointestinal hormones se-
cretin, gastrin, and CCK. The closing of the Old
Frontier was marked by the new developments
which opened the Middle Frontier: advances in
protein chemistry and radioimmunoassay (RIA),
which allowed the ready purification, sequencing,
and synthesis of peptides, and the evaluation of
their status as hormones. This occurred in or
about the late 1960s. Now the flow of scientific
events reversed from the classic approach of Bay-
liss and Starling. No longer did we begin with the
physiologic action, such as the secretion of water
and bicarbonate, and proceed to the identification
of the hormone, for example, secretin. In the new
mode we purify, sequence, and synthesize pep-
tides and then look for their physiologic roles.

The Middle Frontier was thus marked by the
appearance of what Mort Grossman called the
Candidate Hormones (9). The application of RIA
methodology allowed the analysis of hormonal
significance and the rapid characterization of het-
erogeneous molecular forms. But perhaps the
most intriguing concept of the Middle Frontier
was that the peptides could be more broadly dis-
tributed and function by nonhormonal modes.

It was found, for example, that gastric inhib-
itory polypeptide (GIP) is not an enterogastrone,
but probably has a hormonal role as an incretin
and might better be called glucose-dependent in-
sulin-releasing peptide. Vasoactive intestinal
peptide (VIP) was detected in many tissues and is
not a hormone at all but a neuropeptide. But CCK
proved to be most instructive for the emerging
conceptualizations of the Middle Frontier.

At first there was little interest in CCK.
Much more work was done with the closely re-
lated gastrin peptides. The source of gastrin's ap-
peal was in its applicability to the study of acid-
peptic disorders. No such obvious rewards were
available to underwrite the more arduous techni-
cal work required for sensitive and specific mea-
surement of CCK peptides, since no clinical states

Vol. 60 No. 3

NEW FRONTIERS IN GI HORMONES— STRAUS

177

of excessive secretion of pancreatic enzymes, he-
patic bile flow, or gallbladder contractility had
been defined or postulated. But then, in the mid-
dle seventies, CCK was found in the central ner-
vous system (10). As a venerable gastrointestinal
hormone, CCK inspired respect but little enthu-
siasm. Found in a new and unexpected location it
seemed more exciting. Adding to the excitement
were the high concentrations throughout the ce-
rebral cortex and subcortical areas, the abun-
dance of the COOH-terminal octapeptide, the con-
centration of immunoreactivity in synaptosomes,
and the release from synaptosomes according to
paradigms usually employed to evaluate neuro-
nal chemicals purported to have a role in synaptic
physiology.

In a sense the end of the Middle Frontier
brings together the concepts of the Old Frontier.
Nervism was an incomplete understanding of the
regulation of digestive organ function. With the
discovery of hormones it appeared that two rather
disparate systems regulated function: the ner-
vous and endocrine systems. Here at the end of
the Middle Frontier we can see how closely re-
lated these seemingly disparate systems are. The
very same peptide, CCK, is synthesized in both
endocrine cells and neurons and regulates func-
tion both as a hormone and as neurotransmitter.
We understand how venerable the peptides are as
an information transfer system, how many bio-
logically active sequences have been conserved
through evolution, even though these sequences
may have been included in new cell types and
taken on new modes of activity and biological
roles. A messenger may have begun as an auto-
crine or paracrine agent in a simple organism and
then developed hormonal and/or neuronal actions
in more complex forms. This is why the term "gas-
trointestinal hormones" has given way to "regu-
latory peptides." Finally, of course, the tremen-
dous insights into the regulation of gene
expression are a part of the Middle Frontier pe-
riod as well, although the greatest contributions
of molecular biology to this field lie ahead.

The New Frontiers

What of the New Frontiers in gastrointesti-
nal hormones? This is an even more artificial and
speculative construct than my already straining
Old and Middle frontiers, but allow me to make a
fool of myself in public. There will be no discovery

of an essential hormone like insulin in the trea-
sure trove of regulatory molecules found in the
gastrointestinal tract. This would appear evident
from the fact that the removal of virtually the
entire alimentary canal does not require the re-
placement of an essential hormone. Nonetheless,
important regulatory humors will continue to be
discovered, and I expect that the liver will be the
source of some of these. Perhaps a factor regulat-
ing renal cortical blood flow will be found.

The future will certainly include develop-
ment of our knowledge of brain-gut relationships.
This will grow from studies of the interactions of
regulatory peptides acting as paracrine, hor-
monal, and neurotransmitter agents. The ability
to make electrophysiologic recordings of single
visceral afferent fibers, the use of the selective
sensory neurotoxin, capsaicin, refined immuno-
histochemical techniques, and other methods will
facilitate these studies. We will see the definition
of the roles of substance P (SP) and calcitonin-
gene-related-peptide (CGRP) in visceral afferent
neurons. We will see thyrotropin-releasing factor
(TRH) emerge as a player in the vagal regulation
of gastrointestinal function. Corticotropin-releas-
ing factor (CRF) will be extensively studied for its
role in stress-related alterations of gastrointesti-
nal functions. And the interrelationship between
CCK released from the gut as a hormone, and
brain CCK in satiety mechanisms will be worked
out.

Of course, as always in science, the most im-
portant and exciting developments will come as a
complete surprise. That's the fun of it all.

References

1. Bayliss WM, Starling EH. On the causation of the so-

called "peripheral reflex secretion" of the pancreas.
Proc R Soc Lond B 1902; 69:352-353.

2. Bayliss WM, Starling EH. The mechanism of pancreatic

secretion. J Physiol Lond 1902; 28:325-353.

3. Ivy AC, Farrell JI, Lueth HC. A hormone for external

pancreatic secretion. Am J Physiol 1927; 82:27-33.

4. Edkins JS. On the chemical mechanism of gastric secre-

tion. Proc R Soc Lond B 1905; 76:376.

5. Okada S. J Physiol Lond 1914/15; 49:457.

6. Ivy AC, Oldberg. Ann J Physiol 1928; 86:599.

7. Harper AA, Raper HS. J Physiol 1943; 102:115.

8. Jorpes E, Mutt V. Acta Physiol Scand 1966; 66:196.

9. Grossman MI, and others. Candidate hormones of the gut.

Gastroenterology 1974; 67:730-755.
10. Straus E, et al. Immunochemical studies relating to chole-
cystokinin in brain and gut. In: Recent progress in hor-
mone research 1981; 37:447^75.

Current Advances in
Gastrointestinal Immunology

Lloyd Mayer, M.D.

Although the initial descriptions of the inflam-
matory bowel diseases are, in the case of ulcer-
ative colitis, more than 100 years old, and in the
case of Crohn's disease more than 50 years old,
surprisingly little progress has been made toward
a firmer understanding of their pathogenetic
mechanisms. The histopathologic features offer
clear evidence of an active ongoing inflammatory
and immune response. However, the question re-
mains: What is driving these responses? Thus far,
despite grand-scale efforts, no consistent patho-
gen has been isolated from the tissues of patients
with inflammatory bowel diseases, nor have
newer sophisticated technologies yielded evi-
dence for unusual viral or bacterial forms. We are
left, then, in the same predicament facing inves-
tigators studying chronic inflammatory disorders
of other organ systems (rheumatoid arthritis, sar-
coidosis, multiple sclerosis, psoriasis): the search
for initiating agents that trigger the acute in-
flammatory process.

A second dilemma arises in the attempt to
explain the chronicity of these diseases. To ex-
plain the initiation of the disease process is one
thing; to explain its perpetuation is quite an-
other. Few infections are chronic, and those that
are (leprosy, tuberculosis, chronic hepatitis, for
example) evoke a host response that plays a large
role in the expression of disease (as does, for ex-
ample, granuloma formation in tuberculosis). The
other end of the spectrum might suggest that
these series of chronic inflammatory states are

Adapted from the author's presentation at the Dedication of
the Henry D. Janowitz Division of Gastroenterology at the
Mount Sinai Medical Center on December 8, 1992. From the
Department of Clinical Immunology, Mount Sinai School of
Medicine. Address reprint requests to the author at Box 1089,
Mount Sinai Medical Center, One Gustave L. Levy Place, New
York, NY 10029.

reflective of an autoimmune process in which im-
munoregulatory mechanisms have gone awry and
tolerance to self is abrogated.

A common thread in these two dilemmas is
the potential role of the host immune system. The
histopathology of the inflammatory bowel dis-
eases is quite reminiscent of immunologically me-
diated disorders: in ulcerative colitis an Arthus
reaction (type III) with polymorphonuclear leuko-
cyte infiltration, macrophage activation, edema,
and immune complex deposition, and in Crohn's
disease a delayed-type hypersensitivity reaction
(type IV) with granuloma formation and T-cell
and macrophage activation. There is little evi-
dence for a directed immunologic attack in either
disease, with the potential exception of an au-
toantibody directed against an epithelial-cell pro-
tein described by Das. Recent data suggest that
complement activation and deposition does occur
in a pattern consistent with the expression of the
40 kd autoantigen, so that, conceivably, comple-
ment-mediated inflammation may result.

Still, the important question remains: What
is driving the initial response, and why don't con-
ventional regulatory pathways suppress the on-
going inflammation? The answer to that question
may relate to defining the normal mechanisms of
immune regulation in the gastrointestinal tract.
What has become abundantly clear over the past
decade is that the immune system in the gut does
not follow conventional rules. There are unique
populations of cells, such as intraepithelial lym-
phocytes, whose function has not been defined,
although clearly, because they sit at the interface
of the internal and external environment, they
must have some key role in maintaining homeo-
stasis. There is also immunologic nonresponsive-
ness, epitomized by oral tolerance, the antigen-
specific suppression of an immune response that
follows oral feeding. That oral tolerance exists in

178

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

ADVANCES IN GI IMMUNOLOGY— MAYER

179

humans has recently been confirmed; whether it
is altered or exists in the inflammatory bowel dis-
eases remains to be determined.

So where do the regulatory mechanisms in
the mucosal immune system exist? Recent data
from several laboratories suggest that intestinal
epithelial cells (lEC) may play a dual role in mu-
cosal immune responses: as a transporter of IgA
from the lamina propria to the lumen, and as a
major antigen-presenting cell. lEC have been
shown to express products of the major histocom-
patibility locus (class II antigens), which present
small antigenic peptides to antigen receptors on T
cells (resulting in T-cell activation). The lEC,
though, are in a unique environment. Many an-
tigens that travel through the stomach and small
intestine are "processed" by intestinal enzymes
and acid. The requirement for processing by lEC
may be limited, and class II molecules may bind
luminal peptides and transport them to the baso-
lateral surface, where interaction with T cells
(through their antigen receptors) can ensue. Al-
though this preprocessing pathway is advanta-
geous, clearly lEC are capable of taking up mac-
romolecules and processing them into immuno-
genic peptides in their own right.

Thus the lEC can serve as an effective immu-
noregulatory or antigen-presenting cell. The
unique feature here comes in the type of T cell
activated in this interaction. In contrast to con-
ventional antigen-presenting cells (macrophages,
B cells, dendritic cells), where helper (CD4^) T
cells proliferate, normal epithelial cells selec-
tively activate suppressor T cells. The explana-
tion for this dichotomy is a current focus of inves-
tigation, but one of the products of such studies
has been the finding that lEC derived from patients
with IBD fail to activate suppressor cells, but rather
possess many features of conventional antigen-pre-
senting cells, including CD4 ^ T-cell activation, and
enhanced antigen uptake and processing.

The consequences are immediately clear. Ac-
tivation of CD4 ^ helper T cells results in the se-
cretion of cytokines which have far-reaching ef-
fects in and outside the immune system. The
failure to stimulate suppressor cells eliminates
the ability of the mucosal immune system to shut
itself off, setting up a vicious cycle of antigen en-
try, immune activation, cytokine secretion, and
chronic inflammation.

Several recent studies support this model of
inflammatory bowel disease pathogenesis. Two
types of T helper cells, Thj and Thg, which secrete
distinct profiles of cytokines, have been identified
in the mouse and more recently in humans. Th^
cells secrete IL2 and 7-IFN and are thought to be
involved in cell-mediated immune responses or
delayed-type hypersensitivity reactions. Th2 cells
secrete IL-4, IL-5, and IL-6, cytokines which pro-
mote antibody secretion. Using highly sensitive
techniques, investigators have shown that a Th^
profile of cytokines exists in mucosal biopsies
from patients with Crohn's disease (potentially
supporting granuloma formation and macro-
phage and T-cell activation), whereas a Thg pro-
file of cytokines exists in ulcerative colitis — more
an antibody-mediated histologic picture. Al-
though these observations are tantalizing, the
mechanism of this selective activation of Th sub-
sets remains to be determined. Clearly some con-
sideration of genetic predisposition (genetic) re-
sulting in immune dysregulation (immunologic)
following exposure to multiple antigens (environ-
ment) would satisfy many investigators.

Regardless of what the final answers to these
issues turn out to be, these newer areas of inves-
tigation have already opened new vistas for ther-
apeutic intervention. Utilization of combinations
of agents which can modulate initiating, second-
ary, and tertiary effects of immune activation
may promote rapid recoveries and long-term sta-
bilization of the clinical course in these diseases.

Evolution of the Controlled
Clinical Trial

John E. Lennard-Jones, MD, FRCP, FRCS

Historical

The first recorded controlled trial of a treatment
was undertaken by a naval surgeon, James Lind,
in 1746. Scurvy was rife among seamen at that
time because of their poor diet. Lind recognized
that the effect of a treatment can be assessed only
by planned comparison with other measures.
Without such a comparison one does not know if
any apparent effect is actually due to the natural
history of the illness or is an unusual result of
chance. He therefore observed six pairs of sailors,
each pair given a different regimen, including a
pair given sea water and a pair given two oranges
and a lemon daily for 6 days. Those given the
citrus fruit recovered with remarkable speed, and
the result was clearly better than with the other
five regimens. Lind reported these results in 1753

(1) and gradually, as the idea of giving fruit and
vegetables gained ground, scurvy ceased to be the
scourge of seamen on long voyages.

It was to be exactly two hundred years before
the next landmark occurred. In 1946 the British
Medical Research Council set up a trial commit-
tee of 15 members to test the newly discovered
streptomycin in the treatment of tuberculosis.
Among its members was Professor (later Sir) Aus-
tin Bradford Hill, who was to become a major pro-
ponent of the randomized controlled clinical trial

(2) . This first trial was a model of its kind. A
selected group of young patients with bilateral
acute progressive pulmonary tuberculosis was

Adapted from the author's presentation at the Dedication of
the Dr. Henry D. Janowitz Division of Gastroenterology on
December 8, 1992 at the Mount Sinai Medical Center. From
St. Mark's Hospital, London.

Address correspondence and reprint requests to Prof J. E.
Lennard-Jones, 55 The Pryors, East Heath Road, London
NW3 IBP England.

randomized into a group treated with conven-
tional bed rest and another treated in the same
way but with the addition of streptomycin.

The results, published in 1948 (3), showed
that the mortality was less and the radiological
improvement greater among those given the an-
tibiotic, but results were best in the first three
months of treatment, attributed to the fact that
development of drug resistance was noted in the
laboratory, often during the second month. Over
the next few years the tuberculosis trial group
tested combination therapies using streptomycin,
and the newly developed INAH and PAS. By
1955, the results of these trials had established
the principles of drug treatment of tuberculosis
followed ever since.

The controlled trial was introduced into gas-
troenterology by Truelove and Witts, who in 1954
began their controlled trial of cortisone in active
ulcerative colitis (4). The trial fulfilled most of the
criteria required of a modern trial. It asked a sin-
gle clinically important question. The type of dis-
ease, namely patients expected to be admitted to
hospital for at least 6 weeks, was defined. These
patients were randomized into those receiving ac-
tive or dummy tablets; both patients and doctors
were blind to the treatment given. The two
groups were stratified into patients with a first
attack or relapse of disease. Outcome measures
were defined in advance, and these measures were
assessed in the short term at 6 weeks and in the
longer term at 9 months and two years. Lastly, the
results were assessed to determine the probability
that the results observed might be due to chance.

Truelove and Witts went on to compare the
effects of cortisone and corticotropin, a question
taken up again by the group at The Mount Sinai
Hospital in their useful comparison of hydrocorti-
sone and corticotropin, both given intravenously,
for the treatment of acute ulcerative colitis (5).

180

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

CONTROLLED CLINICAL TRIALS— LENNARD-JONES

181

Trial Design

The commonest trial design remains the com-
parison of two groups treated prospectively. The
cross-over design is only appropriate when a
symptom is persistent over a relatively long pe-
riod; it should respond rapidly to an active treat-
ment and recur rapidly when the treatment is
stopped without a carry-over effect from one pe-
riod to the next. For these reasons it is a design of
limited usefulness, though it is an attractive con-
cept. The trial of 6-mercaptopurine in Crohn's dis-
ease reported from Mount Sinai and Lenox Hill
Hospitals (6) was designed in this way and illus-
trates the difficulties of this design. Only 39 of the
patients were crossed over; of the 33 who did not
cross over, 22 refused to do so because they were
well on the first treatment.

A variant on this trial is the so-called N = 1
design, in which repeated comparisons of two
treatments are made in one patient under blinded
conditions and in random order (7).

A sophistication of the usual comparison of
two treatments (A and B) is a factorial design. A
double dummy (a and b) is used, and the patients
are divided into four groups, one of which takes
both test treatments (A and B), each of two groups
takes one test treatment — either (A and b) or (a
and B) — with a dummy identical in appearance
with the other treatment, and the fourth group
takes both dummies (a and b). Thus each test
treatment can be compared against placebo, the
test treatments can be compared with one an-
other, and a combination of both treatments can
be compared with either alone or with the control
group taking placebo. It is an elegant design
which economizes in patient numbers and gives
maximal information, including the possible ad-
ditive effects of two different treatments.

The sequential trial is an interesting design
which aims to give a significant comparison be-
tween two treatments with small numbers or in-
dicate when a significant difference is unlikely to
be observed. Patients are treated in pairs, one
given one of the two test treatments and the other
the alternative. If one patient improves more
than the other, a diagonal is drawn on the graph,
upward if A is better than B or downward for the
converse (Fig, 1). If the results for the pair are
equal, the result is regarded as "tied" and no score
is recorded. If the resulting line on the graph hits
one of the boundaries, it indicates a significant
advantage for one member of the pair. If the line
emerges on the right without crossing a bound-
ary, no significant difference is likely to be gained
by continuing.

In one such trial, in which hydrocortisone re-

SrMPTOMATIC REMISSION

Fig. L Example of a sequential trial. In six successive pairs
of patients, improvement was greater with hydrocortisone en-
emas than with placebo in the treatment of active ulcerative
colitis. When plotted on the graph (thick line) this result in-
tercepts a boundary showing that the result is significant at
the P < 0.05 level. Reproduced with permission of the author
and publisher (ref. 8).

tention enemas were tested against a placebo in
the treatment of distal active ulcerative colitis, a
significant result at the P = 0.05 level (Fig. 1)
was obtained after treatment of only six pairs of
patients (8).

The withdrawal trial is a particularly sensi-
tive design for use when a drug maintains a dis-
ease in remission. In this type of trial, patients
who are well while taking the drug agree to ran-
dom substitution of a dummy tablet for the test

100

L Azathiopnne (n=33)

• •*

• ■*

• • B H

<« ■ ■ 1

Placebo (n=34)

I ' I

X^ = 4 26. p= 0039
I

0 100 200

Days

Fig. 2. Example of a withdrawal trial in 67 patients with
ulcerative colitis who were in remission while taking azathi-
oprine. When a placebo was substituted in half the patients,
the relapse rate was significantly greater over one year in the
group taking placebo than in those who continued to take
azathioprine. Reproduced with permission of the authors and
publisher (ref 9).

182

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

medication in half the group, and the time to re-
lapse is noted among those who continue the drug
and those who stop it. Fig. 2 is an example of such
a trial among patients with ulcerative colitis
whose disease was in remission while taking aza-
thioprine. The relapse rate was greater among
those who stopped the drug than among those
who continued it (9). Previously a therapeutic ef-
fect of azathioprine in ulcerative colitis had been
in doubt. The reason that this type of trial is such
a sensitive test is that patients who do not appar-
ently respond to the drug or those who cannot
take it because of side-effects have been excluded,
leaving only apparent responders in the test
group.

Trial Size

The obvious truth can first be stated: a large
difference between two treatments can be de-
tected by a small trial, detection of a small differ-
ence requires larger numbers.

At the outset it is important to define what is
regarded as a clinically important difference, be-
cause this defines the size of trial required. The
number of patients in the two treatment groups is
determined by the uncertainty inherent in study-
ing a relatively small sample out of a theoretical
large total population. Two errors can be con-
ceived. First, a difference may be observed and
regarded as real when in fact there is no differ-
ence. This is the type 1 error. It is defined by the
conventional test of significance, which is based
on the null hypothesis that there is in fact no
difference. The test of significance estimates the
proportion of trials of the same size in which the
observed deviation from zero would be found.
Thus, if P is calculated to be 0.05, the difference
might be observed in 1 in 20 trials, presuming
that there is in fact no real difference. The null
hypothesis is thus possible but improbable. Re-
peated testing among subgroups in a single trial
may yield a "significant" result in one of 20 tests,
a fallacy to be avoided by defining in advance one
or a few hypotheses to be tested (10).

The type II error starts from a different prem-
ise. It assumes that there may be a real difference
between the treatments and estimates the proba-
bility that this difference would be missed. The
likelihood of missing a difference depends on its
magnitude, the numbers in the trial, and the de-
gree of significance demanded of a positive result.
The power of a trial is the probability that a dif-
ference will not be missed. Thus a power of 0.9
means that the clinically useful difference de-
fined in the trial protocol is likely to be detected
in 9 out of 10 trials.

In modern trials it is essential that the trial
size be determined at the outset, and convenient
tables are available into which may be entered
the likely control value and the difference from
control regarded as clinically important, the de-
gree of significance which would be accepted as
demonstrating the probability that such a differ-
ence exists, and the proportion of trials which
would detect such a difference. The table then
shows the number of patients in each test group
necessary to fulfill these conditions.

If trial size is not defined in the protocol,
when the results are published, the reader does
not know if numbers were arbitrary and de-
pended on the easily available number of cases, if
the trial failed to achieve the numbers planned,
or if the trial was stopped at a particular point
because significance was achieved (10).

If a result is simply reported as "significant'
or "not significant," the reader has no concept of
the magnitude of the difference observed or the
possibility that a real difference exists but was
not detected. It is therefore becoming increas-
ingly common to state the mean difference be-
tween two treatments and the 95% confidence
limits within which the true mean for an infinite
population is expected to lie (11, 12).

Thomas Chalmers and his colleagues, work-
ing at Mount Sinai Medical Center and at the
University of North Carolina, drew vivid atten-
tion to this problem in 1978. They analyzed 71
controlled trials from peer-reviewed journals such
as the New England Journal of Medicine, the
Journal of the American Medical Association, and
the Lancet, which had reported "no difference" be-
tween two treatments (13). It became apparent
that a number of the trials showed a mean differ-
ence which may well have differed from zero,
though the result was "not significant" because
the confidence limits overlapped the zero position.
Among these trials only one third were large
enough to detect a difference of 50% from the con-
trol value in 9 out of 10 trials; the trials were thus
too small and real differences were possibly being
missed. If trials have to be larger they often have
to be multicenter. This necessarily increases pa-
tient variability and observer variation. The re-
sult is reduced sensitivity but wider applicability
of the trial result.

The question of applicability of trial results is
an important one. A trial report should state
what proportion of all eligible patients were in-
cluded. Studies suggest that it is necessary to
screen about twice as many eligible patients as
are actually recruited (14). The results of any
trial are only applicable to patients defined by the

Vol. 60 No. 3

CONTROLLED CLINICAL TRIALS— LENNARD-JONES

183

eligibility criteria. It is therefore also important
to know what proportion of the total population of
patients is represented by those eligible for the
trial.

Outcome Measures

No trial can succeed unless clear outcomes
are defined. These can be subjective and rely on
the judgment of patients or clinicians. Ideally one
needs objective measures which are subject to
minimal bias, such as fever, or no bias, such as
mortality. In practice many apparently objective
criteria — such as healing of ulcers — depend on ob-
server judgment. Intermediate types of outcome
are numerical indices, often a mathematical ex-
pression of a consensus opinion, such as the
Crohn's Disease Activity Index, which is made up
of symptoms, signs, a laboratory value, and use of
symptomatic treatment. In my view, such indices
may give a misleading sense of objectivity, and a
study has also shown that clinicians are bad at
calculating them (15).

There is much to be said for the therapeutic
goals devised by Daniel Present and his col-
leagues for the trial of 6M-P in Crohn's disease
(6). Each goal expresses a real clinical objective
which can be expressed as a change from zero,
either improvement or deterioration on an arbi-
trary ordinal scale. The resulting figures can be
presented as a mean, to give an overall change
with time, or as the values at the end of the trial.
In expressing the results it is important to define
the degree of clinical improvement, which may
vary between slight ( + 1) and complete ( + 3)
achievement of the goal, and not only its statisti-
cal significance.

Ethics

The ethics of controlled trials present major
problems. When is it ethical to use a placebo? If
an active treatment is available, should the new
treatment be compared with it? I have recently
been involved with this problem in a trial of a
fluorinated steroid with low bioavailability
against prednisolone, both given by mouth in the
treatment of chronic active Crohn's disease. The
new steroid proved to be less effective than pred-
nisolone but, because there was no placebo group,
we do not know if the drug is active or not (16).
Since it is free of side effects, this remains an
important question. It is for this reason that the
Food and Drug Administration has rightly en-
couraged placebo-controlled trials of new drugs.
There is no easy answer to this real difficulty, but

clearly trials should include an untreated group
whenever possible.

This raises the matter of informed patient
consent. Doctors are used to dealing with uncer-
tainty, but patients are not. If the outcome of a
trial was known there would be no point in con-
ducting it. Patients therefore have to understand
the uncertainty inherent in the proposed trial be-
fore they can give informed consent. This is diffi-
cult because patients want their doctors to know
what is best for them. The principle of uncer-
tainty requires a change in the attitudes of both
doctors and patients. Doctors need to recognize
their fallibility and patients need to recognize
that the motives of the doctor are good but that
there is uncertainty in the outcome of every rec-
ommendation.

Clearly, the safety of the patient is para-
mount in any projected trial. There must be a
reasonable expectation that the new treatment to
be tested will be safe. Pharmaceutical companies
are now often prepared to compensate patients
financially for any unexpected adverse event due
to the treatment when it is given within the pro-
tocol of a trial initiated by the company (17).

Collaboration in Trials

Collaboration with Patients. The developing
relationship between patients and doctors re-
quires a much greater participation of patients
within the design and conduct of controlled trials
than hitherto. This matter has been brought to
the fore during trials of potential new drugs for
HIV (18). Patients have felt that they might be
denied a possible active treatment for a disease
which is at present incurable if they enter a pla-
cebo-controlled trial. Patient associations are now
becoming involved in the understanding and de-
sign of trials so that their members know that
their interests and welfare are truly being consid-
ered. Issues of quality of life, rather than medical
indices of improvement, are now high on the
agenda of committees planning drug trials. It is to
be hoped that this involvement of patients in clin-
ical trials will become commonplace in many
chronic diseases, including inflammatory bowel
disease.

Collaboration with Industry. A major factor
in most trials nowadays is collaboration with the
pharmaceutical industry. Drug regulatory au-
thorities control the introduction of new drugs or
changed indications for old drugs. The FDA and
equivalent bodies in other countries base their
decision on the efficacy and safety of drugs. So
that their decisions are based on reliable informa-

184

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

o
o

SAS AZA ST

RD=0.13 RD=0.10 RD=033

95%CI=0 03-0 24 95%CI=-0 06-0 25 95%CI=0 09-0 56
COMPLETE CLINICAL REMISSION

Fig. 3. Results of a meta-analysis of controlled trials of sin-
gle drug treatments (black bars! in active Crohn's disease. The
placebo response (diagonal bars) was about 30%. Sulfasalazine
(SAS) and azathioprine (AZA) showed a modest and signifi-
cant therapeutic advantage of about 10%. Steroids (ST)
showed the greatest therapeutic advantage. Reproduced with
permission of the authors and publisher (ref. 21).

tion, they have issued "Guidelines for Good Clin-
ical Practice" in drug trials. Doctors have to deal
with large organizations, there is much extra ad-
ministrative work, and they have to accept super-
vision by external monitors.

Almost gone are the days of clinical indepen-
dence and close personal involvement in a small
trial. All concerned have to recognize the eco-
nomic problems facing a drug firm in the devel-
opment and clinical assessment of a new drug.
Financial rewards may be offered to those under-
taking clinical trials which are subject to their
own ethical constraints.

Elsewhere, I and others have pointed out the
considerable information which can be gained
from small clinical trials using confidence inter-
vals in initial dose-ranging trials (19). Such stud-
ies are to be encouraged before launching a mul-
ticenter trial with its complex organization and
large financial outlay.

thinks of the control of gastrointestinal bleeding,
screening for colorectal cancer in the general pop-
ulation, and symptomatic treatments for func-
tional disorders.

His dictum implies that our trials should be
designed to consider clinically important thera-
peutic outcomes — do patients feel better, can they
function more effectively, do they live longer?
How far does intervention by a doctor really help?
For example, are regular follow-up consultations
useful? We must put our practice under scrutiny.
He also believed that it is unethical not to report
the results of clinical experiments. There is a bias
against publication of trials with a negative re-
sult. It is now suggested that all trials should be
registered, that systematic reviews should be un-
dertaken, and that meta-analysis is a helpful pro-
cedure. The use of standard abstracts would help
this process. An information center to coordinate
the results of controlled trials, named for Coch-
rane, has recently been set up for this purpose in
England. Computer-based sources of information
are being developed in America.

Tribute to Dr. H. Janowitz

Fig. 3 is taken from the most recent study
published by Henry Janowitz and his colleagues,
a meta-analysis of single drug treatments in
Crohn's disease (21). Note the marked placebo re-
sponse of around 30%, the therapeutic advantage
(TA) calculated for 3 drugs, sulfasalazine, azathi-
oprine, and steroids, and the confidence limits of
the relative risk value (similar to TA).

What more could one ask of an analysis of
this type? In 1992, as throughout his career, Dr.
Janowitz has given a lead and influenced world
opinion. It is indeed fitting that the Department
of Gastroenterology at Mount Sinai should be
dedicated to him. I wish Dr. Janowitz, and the
Department, the greatest possible success in the
future.

Controlled Trials Define
Effective Treatments

All over the world, health costs are tending to
rise out of control. Prof A. L. Cochrane in the
United Kingdom enunciated the law that "all
treatment must be proved to be effective" (20).
Treatment in this context can be interpreted
widely to include all types of medical interven-
tion, not only drugs, but surgical procedures, di-
ets, psychological intervention, physiotherapy,
and use of hospital beds. He believed that the use
of randomized controlled trials should permeate
our clinical practice. In gastroenterology, one

References

1. Lind JA. A treatise of the scurvy. Edinburgh: 1753.

2. Doll R. Sir Austin Bradford Hill and the progress of med-

ical science. Br Med J 1992; 305:1521-1526.

3. Medical Research Council. Streptomycin treatment of pul-

monary tuberculosis. Br Med J 1948; 2:769-782.

4. Truelove SC, Witts LJ. Cortisone in ulcerative colitis: fi-

nal report on a therapeutic trial. Br Med J 1955; 2:
1041-1048.

5. Meyers S, Sachar DB, Goldberg JD, Janowitz HD. Corti-

cotropin versus hydrocortisone in the intravenous
treatment of ulcerative colitis: a prospective, random-
ized, double-blind clinical trial. Gastroenterology 1983;
85:351-357.

6. Present DH, Korelitz BJ, Wisch N, Glass JL, Sachar DB,

Vol. 60 No. 3

CONTROLLED CLINICAL TRIALS— LENNARD-JONES

185

Pasternack BS. Treatment of Crohn's disease with
6-mercaptopurine: a long-term, randomized, double-
blind study. N Engl J Med 1980; 302:981-1026.

7. Guyatt G, Sackett D, Taylor DW, Chong J, Roberts R,

Pugsley S. Determining optimal therapy randomized
trials in individual patients. N Engl J Med 1986; 314:
889-892.

8. Watkinson G. Treatment of ulcerative colitis with topical

hydrocortisone hemisuccinate sodium: a controlled trial
employing restricted sequential analysis. Br Med J
1958; 2:1077-1082.

9. Hawthorne AB, Logan RFA, Hawkey CJ, et al. Random-

ised controlled trial of azathioprine withdrawal in ul-
cerative colitis. Br Med J 1992; 305:20-22.

10. Pocock SJ, Hughes MD, Lee RJ. Statistical problems in

the reporting of clinical trials: a survey of three medical
journals. N Engl J Med 1987; 317:426-432.

11. Gardney MJ, Altman DG. Confidence intervals rather

than P values: estimation rather than hypothesis test-
ing. Br Med J 1986; 292:746-750.

12. Bulpitt CJ. Confidence intervals. Lancet 1987; 1:494-497.

13. Freiman JA, Chalmers TC, Smith H Jr, Kuebler RR. The

importance of beta, the type 2 error and sample size in
the design and interpretation of the randomized control
trial: survey of 71 "negative" trials. N Engl J Med 1978;
299:690-694.

14. Charlson ME, Horwitz RI. Applying results of randomised

trials to clinical practice: impact of losses before ran-
domisation, Br Med J 1984; 289:1281-1284.

15. De Dombal FT, Softley A. lOIBD report no 1: observer

variation in calculating indices of severity and activity
in Crohn's disease. Gut 1987; 28:474-^81.

16. Wright JP, Jarnum S, Schaffalitzky de Muckadell O,

Keech ML, Lennard-Jones JE. Oral fluticasone propi-
onate compared with prednisolone in treatment of ac-
tive Crohn's disease. Eur J Gastroenterol Hepatol 1993;
(in press).

17. Diamond AL, Laurence DR. Compensation and drug tri-

als. Br Med J 1983; 287:675-677.

18. Institute of Medical Ethics Working Party. AIDS, ethics,

and clinical trials. Br Med J 1992; 305:699-701.

19. Powell-Tuck J, MacRae KD, Healy MJR, Lennard-Jones

JE, Parkins RA. A defence of the small clinical trial:
evaluation of three gastroenterological studies. Br Med
J 1986; 292:599-602.

20. Cochrane AL. Effectiveness and efficiency; random reflec-

tions on health services. Cambridge: British Medical
Journal and The Nuffield Provincial Hospitals Trust,
1989.

21. Salomon P, Kornbluth A, Aisenberg J, Janowitz HD. How

effective are current drugs for Crohn's disease? A meta-
analysis. J Clin Gastroenterol 1992; 14:211-215.

Advances in Knowledge of
Inflammatory Bowel Disease at
Mount Sinai Medical Center

Daniel H. Present, M.D.

Dr. Henry D. Janowitz wrote in his book, In-
flammatory Bowel Disease: A Personal View (1),
"Ever since I met Burrill Crohn, Leon Ginzburg
and Gordon Oppenheimer when I came to The
Mount Sinai Hospital in New York in 1939, I
have had the sense of participating in the exciting
accumulation of knowledge of inflammatory
bowel disease and to have been at the frontiers of
both ignorance and uncertainty as well as in-
creasing knowledge."

I personally have had the same emotions
since I came to Mount Sinai in 1961 and as a
gastroenterology fellow from 1964 to 1966. 1 have
been the beneficiary of a tradition of academic
excellence and have been exposed to world-fa-
mous inflammatory bowel disease experts, in-
cluding Drs. Janowitz, John Garlock, Leon Ginz-
burg, Richard Marshak, Arthur Aufses, Jr., Al-
bert Lyons, and Burton Korelitz. I have also
worked with outstanding contemporaries who are
experts in IBD, including David Sachar, Sam
Meyers, Irwin Gelernt, Adrian Greenstein, and
Lloyd Mayer. This tradition makes discussing
knowledge of inflammatory bowel disease gener-
ated at Mount Sinai Medical Center a monumen-
tal task, since if I were simply to list each paper
written at this institution on IBD the list would
stretch to encyclopedic proportions. For my retro-
spective review here, I therefore asked my col-
leagues to recommend what they thought were

Address reprint requests to the author, Clinical Professor
of Medicine, Mount Sinai School of Medicine, at 12 East 86th
Street, New York, NY 10028.

186

the outstanding accomplishments and papers
published in inflammatory bowel disease at
Mount Sinai. I tabulated the results to see if they
agreed with my own personal prejudices, and for
the most part they did.

Historical Perspectives. It is interesting to
speculate about the origins of knowledge of ulcer-
ative colitis and Crohn's disease. In prior centu-
ries, nonspecific ulcerative colitis was likely
confused with infectious colitis. The name "ulcer-
ative colitis" was first used by Sir Samuel Wilks
in 1859, and Wilks and Moxan published a classic
description of ulcerative colitis in 1875 (2),
slightly over 117 years ago. On the other hand,
Crohn's disease appears to be a more recently rec-
ognized entity. It was not specifically identified
by the great pathologists in Europe in the 19th
century, although references have been cited as
early as 1806. Wilks and Moxan also described
acute ileitis with a pathological picture that
sounds similar to what we know as Crohn's dis-
ease: "We have met with severe local ileitis in the
shape of thickening of the whole of the coats, the
wall thick with inflammatory lymph found in
patches of from 6 inches to 2-3 feet."

A Scottish surgeon, Dalziel, in 1913 de-
scribed chronic interstitial enteritis (3) in the
same terms as classical descriptions of Crohn's
disease, but no followup reports were forthcom-
ing. Many advocates of Dalziel, who claim his pa-
per as the first description, blame this loss of rec-
ognition of his work on the start of World War I.
In the 1920s a paper by Moschkowitz and Wilen-
ski (4), describing nonspecific granulomas of the
intestine, and a paper by Moschkowitz entitled
"Sarcoidosis of the Colon" detailed what we can
retrospectively identify as typical Crohn's ileo-
colitis or colitis. These studies separated this en-

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

IBD AT MOUNT SINAI— PRESENT

187

tity from tuberculosis, lymphoma, diverticulitis,
and other disorders.

Several years later, Drs. Leon Ginzburg and
Gordon Oppenheimer, continuing the work of
Moschkowitz and Wilenski, utilizing the pathol-
ogy provided by Dr. A. A. Berg's surgical service,
accumulated 12 cases of localized ileitis in the
terminal ileum. Dr. Burrill Crohn had two pa-
tients with similar disease who also underwent
surgery by Dr. A. A. Berg, and the details of these
14 patients were compiled to describe a new en-
tity which was presented at the annual meeting
of the American Medical Association. The three
authors of this paper called the entity "terminal"
ileitis, but according to Dr. Marshak (5), Dr. J. A.
Bargen protested that the term was bad since it
suggested a fatal illness, and Dr. Crohn agreed to
change the name to "regional" ileitis. This paper
by Drs. Crohn, Ginzburg, and Oppenheimer (6) is
certainly one of the major achievements in in-
flammatory bowel disease at Mount Sinai.

Prior reports had been made of the same
granulomatous process present in the colon, al-
though it was originally thought that the disease
stopped at the ileocecal valve. However, Dr.
Ralph Colp in 1934 reported a case with ileocecal
involvement showing that the disease could cross
the valve (7).

Successes have occasionally been tempered
by disappointments. A significant one was the
failure of otherwise competent pathologists at
Mount Sinai to recognize that Crohn's disease
could affect the colon; in reports on the surgically
resected specimens, they were typically described
as regional ileitis and ulcerative colitis. It re-
mained for Lockhart-Mummery and Morson (8) to
clearly define, in 1960, the clinical and patholog-
ical features of Crohn's or granulomatous colitis.
These authors clearly separated this entity from
ulcerative colitis. However, to our credit, our ra-
diological colleagues Marshak and Wolf (9) were
quick to describe the x-ray features of what was
then called granulomatous colitis. In fact, in a
paper before Lockhart-Mummery and Morson's
report, Marshak, Wolf, and Eliasoph described
segmental colitis (10). However, because of the
pathological reports noted above, they failed to
distinguish segmental Crohn's disease from seg-
mental ulcerative colitis.

Dr. Richard Marshak joined Dr. Crohn's of-
fice in 1946 as his radiologist, and Marshak sub-
sequently became Dr. Janowitz's radiologist. He
had the opportunity to see thousands of cases of
IBD, and he presented a sequential radiographic
history of Crohn's disease distinguishing the non-
stenotic from the stenotic phase (5).

The tradition of documenting the features of
inflammatory bowel disease has been carried on
in recent years in endoscopy by one of the true
leaders in this field. Dr. Jerome D. Waye, an out-
standing teacher, clinician, and educator, has
taught many gastroenterologists how to distin-
guish ulcerative colitis from Crohn's disease as
well as other inflammatory colitides (11).

Megacolon. A major achievement among the
radiological contributions from Mount Sinai is a
paper on toxic dilation of the colon in the course of
ulcerative colitis (12) that reported on 19 episodes
in 16 patients, stressing the radiographic features
of diagnostic and prognostic significance. In a ret-
rospective review there was at least one patient
in this group who had Crohn's disease. This was
therefore the first description of megacolon in
Crohn's disease, but once again, the pathologic
report was that of severe ulcerative colitis with
diffuse ileitis. The authors describe cecostomy as
successful in some patients and anticipated the
future blow-hole surgical technique of Turnbull
(13).

This basic paper stimulated the interest of
Mount Sinai physicians for several decades.
Other contributions to knowledge of this particu-
lar complication of inflammatory bowel disease
include the papers of Meyers and Janowitz (14) on
the role of steroids in megacolon and the large
series reported by Adrian Greenstein et al. (15) on
toxic megacolon in 75 of 1236 IBD patients, a re-
port that summarizes the institution's experi-
ence. My group's contribution to the medical
management of toxic megacolon (16) demon-
strated decompression in all our patients using a
comprehensive medical treatment plan plus the
new addition of a simple rolling technique which
allowed gas to move from the most anterior por-
tion of the colon (transverse) and to be subse-
quently expelled. The latter technique was dis-
covered serendipitously when a patient of Dr.
Janowitz with megacolon rolled on his abdomen
to pick up a book that had fallen on the floor and
promptly relieved himself of air and stool across
the room, and I was — how shall I say it — fortu-
nate enough to walk into the room 15 seconds
later; I subsequently received credit for describ-
ing the rolling technique.

Surgical Treatment. Crohn's disease
achieved great notoriety when President Dwight
D. Eisenhower underwent surgery for obstruction
in 1956 and Crohn's disease became a household
word. There was great debate postoperatively
whether the correct surgical procedure had been
performed (17); the president had an ileotrans-
verse colostomy in continuity, which was not uni-

188

THE MOUNT SINAI JOURNAL OF MEDICINE

May 199J

versally accepted as the preferred surgical proce-
dure. Many of the debated facts were based on
reports from the Mount Sinai Hospital on the sur-
gical management of Crohn's disease.

The initial report in 1945 by Garlock and
Crohn (18) of 164 cases detailed the results of sur-
gery in the treatment of regional enteritis; this
report was updated in 1951 (19). These long-term
followups contributed to the debate on the best
surgical procedure for this disorder. Dr. Garlock
also published a paper relating his experience in
the surgical treatment of ulcerative colitis (20).
Following in this great surgical tradition of A. A.
Berg, Ginzburg, and Garlock is the comprehen-
sive monograph by Arthur Aufses, Jr., "The Sur-
gery of Granulomatous Inflammatory Bowel Dis-
ease" (21). Dr. Aufses has published extensively
and taught several generations of surgeons in the
operating room, at the bedside, and in lecture
halls, and he has an extensive bibliography in
this field. I make particular mention of Dr. Aufses
because I learned from his advice, and now per-
sonally recommend, that gastroenterology fellows
go into the operating room to see what surgery in
these patients entails, since only in this way can
they fully understand the clinical and pathologi-
cal correlations of inflammatory bowel disease.
Dr. Aufses has trained many surgeons in techni-
cal skills as well as clinical excellence, resulting
in outstanding publications (22-23) from a group
including Drs. Isidore Kreel, Irwin Gelernt, Joel
Bauer, and Barry Salky, and especially Dr.
Adrian Greenstein, as well as many others.

Many would agree with me that one of the
major contributions in the postoperative manage-
ment of inflammatory bowel disease arose at The
Mount Sinai Hospital. The first reported ileosto-
my group in the world was started at Mount Sinai
about 1950-1951 and is described in the form of a
letter to the Journal of the American Medical As-
sociation by Dr. Albert Lyons (24). The club was
named the QT Alumni, not to suggest that an
ostomy had to be kept "on the QT," but rather
after ward Q, the male surgical ward, and ward T,
the female surgical ward. The club then became
The QT New York and, eventually, the Ileostomy
Association QT — New York. A colostomy group
was also formed, and Dr. Lyons and Dr. George
Schreiber arranged an all-day convention, held at
the New York Academy of Medicine and attended
by doctors and patients representing ostomy
groups from the immediate New York area. It
was decided at this meeting to try to create a na-
tional organization; ultimately the United Os-
tomy Association came into existence. From
small beginnings in the QT Club, incalculable

help in medical education and psychological ad-
justment has been generated for patients anc
families. This association has enabled patients tc
return to normal lifestyles and has substantial!)
enhanced quality of life for many. Most observers
account it a major contribution.

Natural History. Studies of the natural his
tory of inflammatory bowel disease have pro
duced an enormous array of original papers thai
have benefited medical care. Recurrent diseas(
has been studied from many aspects. Dr. Burtor
Korelitz in 1972 refuted the prior literatun
which stated that after total colectomy and ileos
tomy for Crohn's disease there was no recurrenc(
in the small bowel. His series (25) noted a 469i
recurrence rate which was subsequently con
firmed at other major medical centers. Greensteir
and Sachar et al. (26) published a major study or
reoperation and recurrence in Crohn's disease
using both crude and actuarial life-table methods
in 160 patients. The authors cited the deficiencies
in the prior literature, requesting more rigid def
initions of disease. They noted the inexorable ten
dency of Crohn's ileocolitis to require repeated op
erations. Subsequently, the same authors hav(
been enthusiastic about reporting the evidence
for two clinical forms of Crohn's disease, a perfo
rating and a nonperforating group, and the neec
for more frequent surgery in the former (27).

A major natural history paper is the study o
700 Mount Sinai patients describing the extrain
testinal complications of both Crohn's disease anc
ulcerative colitis (28). In this classic study, com
plications were classified as related either to coli
tis or to the small bowel or as nonspecific. This
review is basic reading for any student of inflam
matory bowel disease, at any level of education
who seeks to understand the mechanisms anc
pathophysiology of such complications as arthri
tis, erythema nodosum, pyoderma, gallstones
and renal stones.

The process of understanding the natural his
tory of inflammatory bowel disease continues ai
Mount Sinai in recent papers. To cite some exam
pies: the first case reports on amyloidosis came
from the Mount Sinai Departments of Radiologj
and Pathology, and in 1960 Dr. Lawrence Wer
ther expanded this experience with an outstand
ing publication (29). In 1992 Greenstein et al
published "Amyloidosis and Inflammatory Bowe
Disease: A 50-year Experience with 25 Patients'
(30), work which now allows for analysis of the
clinical patterns of this serious complication.

Urological complications have been high or
the natural history list, starting with Ginzburg
and Oppenheimer's paper (31). This initial de-

Vol. 60 No. 3

IBD AT MOUNT SINAI— PRESENT

189

scription of urological complications has been fol-
lowed by papers on ureteral obstruction (32), hy-
dronephrosis (33), and urolithiasis (34).

Clinical Trials. Among several significant
therapeutic trials performed at The Mount Sinai
Hospital is the study of Sam Meyers and others
comparing ACTH and hydrocortisone in the
treatment of severe ulcerative colitis (35). The
study demonstrated clearly that intravenous
ACTH was the drug of choice for severe ulcerative
colitis if the patient had not received prior corti-
costeroids, whereas intravenous hydrocortisone
was preferable for those patients already receiv-
ing steroid therapy. The long-term followup of
these patients also provided valuable data on the
effect of steroid therapy on the natural history of
the ulcerative colitis.

Sam Meyers also led a group in a double-
blind controlled trial using olsalazine in the
treatment of ulcerative colitis patients who were
intolerant of sulfasalazine (36). This is an often-
quoted trial in that we not only showed efficacy
for active disease but at the same time demon-
strated a dose response with this therapeutic
agent. Our trial helped to obtain Food and Drug
Administration approval for the first new 5- ASA
agent introduced in the United States. Further
international studies of olsalazine have con-
firmed its efficacy, not only in active treatment,
but also in prevention of relapses in ulcerative
colitis (see, for example, 36A).

A major contribution has been the demon-
stration of the efficacy of immunosuppressives in
the treatment of inflammatory bowel disease.
Burt Korelitz, David Sachar, Nathaniel Wisch,
Bernard Pasternack, and I reported on a double-
blind controlled trial using 6-mercaptopurine in
the treatment of Crohn's disease (37). This has
been the basis for multiple followup studies, in-
cluding evidence for the efficacy of this agent in
chronic disease, the only controlled trial in the
treatment of fistula (38), and providing the evi-
dence and the rationale for the failure of the na-
tional cooperative trial, which used azathioprine
but stopped the trial prematurely (39).

Finally, the authors in followup have docu-
mented the limited toxicity of long-term use of
6-MP in the treatment of inflammatory bowel dis-
ease (40). This trial and subsequent studies have
resulted in the avoidance of surgery with main-
tenance of a high quality of life in innumerable
patients suffering from inflammatory bowel dis-
ease. Throughout the world more and more pa-
tients are being treated with these agents.

Work along these lines continues at Mount
Sinai; under the leadership of Simon Lichtiger, a

double-blind controlled trial has demonstrated
the efficacy of cyclosporine in ulcerative colitis
(41; see also unpublished data: Lichtiger S,
Present DH, Kornbluth A, et al. A placebo con-
trolled double-blind randomized trial of cyclospor-
ine in severe steroid-refractory ulcerative colitis),
the first new therapy for this disease in over 35
years.

Cancer and IBD. Leon Ginzburg gave the
first description of carcinoma in the small bowel
in a patient with Crohn's disease (42), and a sub-
sequent important paper by Adrian Greenstein
and David Sachar, the current chief of the Mount
Sinai Division of Gastroenterology, (43) reported
on the risks of cancer in ulcerative colitis and the
factors determining risk in 267 patients. The
same group has reported on the increased risk of
extraintestinal cancers in inflammatory bowel
disease patients (44).

Basic Science. Although Mount Sinai has a
rich clinical tradition, basic science is not ig-
nored, especially with the addition of Lloyd
Mayer to our faculty. Dr. Mayer is considered one
of the world leaders on basic immunologic defects
in patients with inflammatory bowel disease. His
data, published in 1990 in the Journal of Clinical
Investigation (45), suggests an intrinsic defect in
epithelial cells taken from IBD patients, result-
ing in the inability to normally stimulate sup-
pressor T cells.

Legacy of Henry D. Janowitz. Among Henry
D. Janowitz's recent interests are his return to
the natural history of Crohn's disease, performing
a critical review of the placebo response in mul-
tiple trials (46), as a recent meta-analysis, con-
ducted with three young fellows, of the efficacy of
drugs currently being used to treat Crohn's dis-
ease (47). Nor should we ignore his important pa-
per on the quality of life after surgery for Crohn's
disease (48).

A major achievement of Dr. Henry Janowitz
is to have been the medical founder, along with
two lay founders, Irwin Rosenthal and William
Modell, of the Crohn's and Colitis Foundation of
America, which began at Mount Sinai as the
Foundation for Research in Ileitis. It subse-
quently became the National Foundation for Ile-
itis and Colitis and is now called the Crohn's and
Colitis Foundation of America. Their initial grant
to Mount Sinai for the first year was $25,000, (of
which only $20,000 was spent) and it enabled us
to launch our immunosuppressive trial as well as
several other studies. Now, 25 years later, the
Foundation has given $21 million to research,
$2.55 million in 1992, including 200 research
grants and 156 training awards. Sixty-one per-

190

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

cent of Foundation recipients go on to receive Na-
tional Institutes of Health funding. In addition,
1700 physicians are now members of the organi-
zation, as are over 100,000 patients and donors.

What a legacy from this remarkable man. In
much of what I have described as Mount Sinai
Medical Center accomplishments in inflamma-
tory bowel disease. Dr. Henry D. Janowitz has
played a major role: as a basic physiologist, a col-
laborator, a teacher, a mentor, and, finally, co-
founder of a major foundation dedicated to sup-
porting research and education internationally.
The name of this outstanding physician and gen-
tleman is a fitting endowment for the Dr. Henry
D. Janowitz Division of Gastroenterology.

References

1. Janowitz HD. Inflammatory bowel disease: a personal

view. St. Louis: Mosby Yearbook 1985.

2. Wilks S, Moxan W. Lectures on pathological anatomy,

2nd ed. London: Churchill, 1875.

3. Dalziel AM. Chronic interstitial enteritis. Br Med J 1913;

2:1068-1070.

4. Moschkowitz E, Wilenski AO. Nonspecific granulomata of

the intestine. Am J Med Sci 1923; 166:48-66.

5. Marshak RH. Granulomatous disease of the intestinal

tract (Crohn's disease). Radiology 1975; 114:3-22.

6. Crohn BB, Ginzburg L, Oppenheimer GD. Regional ileitis:

a pathologic and clinical entity. JAMA 1932; 99:1323-
1329.

7. Colp R. A case of nonspecific granuloma of the terminal

ileum and the cecum. Surg Clin N Am 1934; 14:443-
449.

8. Lockhart-Mummery HE, Morson BC. Crohn's disease (re-

gional enteritis) of the large intestine and its distinc-
tion from ulcerative colitis. Gut 1960; 1:87-105.

9. Wolf BS, Marshak RH. Granulomatous colitis (Crohn's

disease of the colon): roentgen features. Am J Roent-
genol 1962; 88:662-670.

10. Marshak RH, Wolf BS, Eliasoph J. Segmental colitis. Ra-

diology 1959; 73:707-716.

11. Waye JD. Endoscopy in inflammatory bowel disease. Clin

Gastroenterol 1980; 9:279-296.

12. Marshak RH, Korelitz BI, Klein SH, et al. Toxic dilation

of the colon in the course of ulcerative colitis. Gastro-
enterol 1960; 38:165-180.

13. Turnbull RB, Jr., Hawk WA, Weakley FL. Surgical treat-

ment of toxic megacolon; ileostomy and colostomy to
prepare patients for colectomy. Am J Surg 1971; 122:
325-331.

14. Meyers S, Janowitz HD. The place of steroids in the ther-

apy of toxic megacolon. Gastroenterol 1978; 75:729-
731.

15. Greenstein AJ, Sachar DB, Gidas A, et al. Outcome of

toxic dilatation in ulcerative and Crohn's colitis. J Clin
Gastroenterol 1985; 7:137-144.

16. Present DH, Wolfson D, Gelernt IM, et al. Medical decom-

pression of toxic megacolon by "rolling." J Clin Gastro-
enterol 1988; 10:485^90.

17. Heaton LD, Ravdin IS, Blades B, Whelan TJ. President

Eisenhower's operation for regional enteritis: a footnote
to history. Ann Surg 1964; 159:661-666.

18. Garlock JH, Crohn BB. An appraisal of the results of sur-

gery in the treatment of regional ileitis. JAMA 1945;

127:205-208.

19. Garlock JH, Crohn BB, Klein SH, Yarnis H. An appraisal

of the long-term results of surgical treatment of re-
gional ileitis. Gastroenterol 1951; 19:414-^23.

20. Garlock JH. Surgical treatment of ulcerative colitis. Dis

Colon Rectum 1959; 2:529-533.

21. Aufses AH, Jr. The surgery of granulomatous inflamma-

tory bowel disease. In: Current problems in surgery.
Chicago: Yearbook Medical Publishers, 1983.

22. Gelernt IM, Bauer JT, Kreel I. Reservoir ileostomy: early

experience with 54 patients. Ann Surg 1977; 185:179-
184.

23. Bauer JT, Gelernt IM, Salky B, et al. Sexual dysfunction

following proctocolectomy for benign disease of the co-
lon and rectum. Ann Surg 1983; 197:363-367.

24. Lyons AS. An ileostomy club. JAMA 1952; 150:812-«13.

25. Korelitz BI, Present DH, Alpert L, et al. Recurrent re-

gional enteritis after ileostomy and colectomy for gran-
ulomatous colitis. N Engl J Med 1972; 287:110-115.

26. Greenstein AJ, Sachar DB, Pasternack BD, Janowitz HD.

Reoperation and recurrence in Crohn's colitis and ileo-
colitis: crude and cumulative rates. N Engl J Med 1975;
293:685-690.

27. Greenstein AJ, Lachman P, Sachar DB, et al. Perforating

and nonperforating indications for repeated operations
in Crohn's disease: evidence for two clinical forms. Gut
1988; 29:588-592.

28. Greenstein AJ, Janowitz HD, Sachar DB. The extraintes-

tinal complications of Crohn's disease and ulcerative
colitis: a study of 700 patients. Medicine 1976; 55:401-
411.

29. Werther JL, Schapira A, Rubinstein O, Janowitz HD. Am-

yloidosis in regional enteritis. Am J Med 1960; 29:416-
423.

30. Greenstein AJ, Sachar DB, Nannan Panday AK, et al.

Amyloidosis and inflammatory bowel disease: a fifty-
year experience with 25 patients. Medicine 1992; 71:
261-270.

31. Ginzburg L, Oppenheimer GD. Urological complications

of regional ileitis. J Urol 1948; 59:948-952.

32. Goldman HJ, Glickman SI. Ureteral obstruction in re-

gional enteritis. J Urol 1962; 88:616-620.

33. Present DH, Rabinowitz JG, Banks PH, Janowitz HD. Ob-

structive hydronephrosis: a frequent but seldom recog-
nized complication of granulomatous disease of the
bowel. N Engl J Med 1969; 280:523-528.

34. Deren JJ, Porush JG, Leavitt MF, Khilnani MT. Nephro-

lithiasis as a complication of ulcerative colitis and re-
gional enteritis. Ann Int Med 1962; 56:843-853.

35. Meyers S, Sachar DB, Goldberg JD, Janowitz HD. Corti-

cotropin versus hydrocortisone in the intravenous
treatment of ulcerative colitis. Gastroenterol 1983; 85:
351-357.

36. Meyers S, Sachar DB, Present DH, Janowitz HD. Olsala-

zine sodium in the treatment of ulcerative colitis among
patients intolerant of sulfasalazine. Gastroenterol
1987; 93:1255-1262.
36A. Courtney MG, Nunes DP, Bergin CF, O'Driscoll M,
Trimble V, Keeling PWN, Weir DG. Randomised com-
parison of olsalazine and meslazine in prevention of re-
lapses in ulcerative colitis. Lancet 1992; 339:1279-
1281.

37. Present DH, Korelitz BI, Wisch N, et al. Treatment of

Crohn's disease with 6-mercaptopurine: a long-term
randomized double blind study. N Engl J Med 1980;
302:981-987.

Vol. 60 No. 3

IBD AT MOUNT SINAI— PRESENT

191

38. Korelitz BI, Present DH. The favorable effect of 6-mercap-

topurine on the fistulae of Crohn's disease. Dig Dis Sci-
ence 1985; 30:58-64.

39. Korelitz BI, Present DH. Shortcomings of the National

Crohn's Disease Study: the exclusion of azathioprine
without adequate trial. Gastroenterol 1981; 80:193-
194.

40. Present DH, Meltzer SJ, Krumholz MP, et al. 6-Mercap-

topurine in the management of inflammatory bowel
disease — short and long term toxicity. Ann Int Med
1989; 111:641-649.

41. Lichtiger S, Present DH. Preliminary report: cyclosporine

in the treatment of severe active ulcerative colitis. Lan-
cet 1990; 336:16-19.

42. Ginzburg L, Schneider KM, Dreizin DH, Levinson C. Car-

cinoma of the jejunum occurring in a case of regional
enteritis. Surgery 1956; 39:347-351.

43. Greenstein AJ, Sachar DB, Smith H, et al. Cancer in uni-

versal and leftsided ulcerative colitis: factors determin-
ing risk. Gastroenterol 1979; 77:290-294.

44. Greenstein AJ, Gennuso R, Sachar DB, et al. Extraintes-

tinal cancers in inflammatory bowel disease. Cancer
1985; 56:2914-2921.

45. Mayer L, Eisenstadt D. Lack of production of suppressor T

cells by intestinal epithelial cells from patients with
inflammatory bowel disease. J Clin Invest 1990; 86:
1255-1260.

46. Meyers S, Janowitz HD. "Natural history" of Crohn's dis-

ease. An analytical review of the placebo lesson. Gas-
troenterol 1984; 87:1189-1192.

47. Salomon P, Kornbluth A, Aisenberg J, Janowitz HD. How

effective are current drugs for Crohn's disease? A meta-
analysis. J Clin Gastroenterol 1992; 14:211-215.

48. Meyers S, Walfish JS, Sachar DB, et al. Quality of life

after surgery for Crohn's disease: a psychosocial survey.
Gastroenterol 1980; 78:1-6.

Reminiscences
of Henry D.
Janowitz

A journey of learning, accomplishment and fulfillment — Henry
D. Janowitz 1959-1963: I arrived at The Mount Sinai Hospital on
June 21, 1959, shortly after my discharge from the Israeli Air
Forces. This was my first trip abroad and needless to say, I was
overwhelmed by my impressions of New York and of The Mount
Sinai Hospital. After two and a half years of active service as a
flight surgeon in the Israeli Air Forces, a position considered to be
most stressful and requiring alertness, agility, and ability for
quick and rational response, I could hardly believe the pace of
activity and sense of urgency everywhere around me in the hospi-
tal. It was a beehive of research and clinical activities, and I was
most impressed and flattered by the opportunity of meeting a num-
ber of famous personalities whom I knew only from their publica-
tions.

I was introduced to Henry in one of the hospital elevators by
Dr. David Adlersberg, who was responsible for my coming to
America and with whom I was supposed to do research on the
differences in the metabolism of lipid emulsions administered in-
travenously or after absorption by the intestine. Henry was quick
to point out that anything that has to do with the intestine should
be of interest to him, that I would be more than welcome to join the
activities of his group, and that I should feel free to discuss ques-
tions or problems related to my research. At that time neither
Henry nor I realized the importance of this statement.

Not long after this conversation. Dr. Adlersberg succumbed to
a massive heart attack, and as a consequence my research project
and my future at The Mount Sinai Hospital became questionable.
By a cruel stroke of fortune, I found myself without financial sup-
port, laboratory facilities, research supervisor, or mentor. The only
one that I could think of turning to was Henry, and his response to
my predicament determined my destiny and reshaped my future.

Henry made sure that I could continue my research, provided
advice and leadership, and also allowed me to participate in the
clinical activities of his department and obtain my clinical training
in gastroenterology. I was given a corner in the laboratories of Dr.
Harry Sobotka, the director of Clinical Biochemistry, who had
done pioneering work on the biochemistry of bile salts. Dr. Sobotka
also gave me the key of a not-well-known storage area several
floors under the basement, literally in the "guts" of the hospital.
Here I found all sorts of glassware and incubation baths and a
micropipette which I used in my work on free fatty acids. In this
area were also stored some of the personal possessions of renowned
refugee scientists from Germany. Here I found the kymograph
used by Dr. Otto Loewi in the discovery of acetylcholine, a discov-
ery for which he was awarded the Nobel Prize in 1936. Here was
also stored a harpsichord which belonged to Dr. Sobotka's wife,
herself an accomplished concert performer, on which I took the
liberty of playing during some of the lonely hours away from the
laboratory bench or ward duties.

The next spring I had the privilege of having one of my ab-
stracts selected for presentation at the meetings in Atlantic City.
I remember the endless rehearsals along the Garden State Park-
way of my presentation with Henry's tenacious coaching on voice,
intonation, body posture, and control of the motions and position of
my arms and hands. It was indeed a miracle that we arrived safely,
because I was also doing the driving. Henry insisted on my re-

192

The Mount Sinai Journal of Medicine Vol, 60 No. 3 May 1993

Vol. 60 No. 3

REMINISCENCES

193

hearsing the answers to a number of possible
questions from the audience and — typical of Hen-
ry's foresight — most of these questions were in-
deed asked after my presentation. The presenta-
tion was a success, and what I learned from
Henry during our drive to the conference proved
to be of immeasurable help in many more presen-
tations in the years to come. This was a journey of
learning.

A few years later, when I accumulated a rea-
sonable list of publications, Henry nominated me
for membership to the prestigious American So-
ciety for Clinical Investigation. Dr. Jules Hirsh of
The Rockefeller University, New York, and Dr.
Marvin Sleisenger of the University of California
School of Medicine, San Francisco, seconded the
nomination. My acceptance as a member of the
"Young Turks" signified the recognition of my re-
search accomplishments and once again rein-
forced for me the immeasurable help and guid-
ance so generously provided by Henry. This was a
journey of accomplishment.

Finally, the day came when I had to start the
journey on my own. I had finished my clinical
training while carrying out active and productive
research, and it was time for me to move on. Al-
though Henry was very interested in having me
join the Division of Gastroenterology at The
Mount Sinai Hospital, he was understanding of
my needs for my professional future and was not
only encouraging and supportive, but also ac-
tively promoted my search for a suitable aca-
demic position. I was appointed chairman of the
Division of Gastroenterology at McGill Univer-
sity, and Director of Gastroenterology at the
Royal Victoria Hospital, Montreal. This was a
journey of fulfillment.

Looking back, I can only wonder what might
have happened if Henry had not been there when
I was in greatest need of help, advice, and guid-
ance. I am glad that Henry was there on that
fateful elevator ride, for outside of my immediate
family there is no one whom I respect more or to
whom I owe so much. It has been a most enjoyable
journey, and I hope that we will have many more
and longer distances to travel together.

Jacques I. Kessler, MD, FRCP(C)
Prof, of Medicine, McGill University
Montreal, Canada

The British Medical Research Council awarded
me an Eli Lilly Travelling Fellowship to spend
1961-62 in a gastroenterology department in the

United States. I took advice, was told that there
were three distinguished gastroenterological cen-
ters in America, and after discreet enquiries ap-
plied to Dr. Janowitz, the newly appointed chief of
gastroenterology at The Mount Sinai Hospital.

He accepted me provisionally by correspon-
dence but asked to interview me when he and Dr.
David Dreiling came to London in June 1961 for
the CIBA Foundation Symposium on the Exo-
crine Pancreas. I told them what I had been do-
ing— in my research 1959-1961 in a medicine-
gastroenterology residency at the Middlesex
Hospital, London, I measured acid output with
Kay's augmented histamine test — and they asked
what I would like to do in my year with them. Did
I know anything about pancreatic secretion? No, I
said. Their response: Then you'll learn about it
when you come!

Maximal Outputs of Gastric Acid and Pan-
creatic Bicarbonate. In London I had devised a
new methodology for calculating maximal acid
output. What I called peak acid output (PAO) was
the mean of the two highest consecutive collection
fractions after a single parenteral dose of agonist

(1) . This PAO was highly repeatable and clearly
represented parietal cell mass. I propounded the
still-current threshold model of duodenal ulcer
disease (DU) in which DU was found only with
PAO more than 15 mmol/h or 10^ parietal cells

(2) .

In New York I joined the late Claude Perrier
of Geneva in trying to establish a maximal alka-
line output of the dog pancreas analogous to max-
imal acid output of the stomach. Our attempt was
feasible and safe only because of the commercial
availability of the more purified Jorpes-Mutt Vit-
rum secretin. By cannulating the major pancre-
atic duct in dogs with long-term Thomas fistulas
we achieved maximum bicarbonate output with
either single intravenous injections or continuous
intravenous infusions, and we could produce in-
hibition by supramaximal doses (3).

We showed that the dog pancreas could se-
crete about one third as much alkali, milliequiv-
alent per milliequivalent, as its stomach could se-
crete acid (4). There was, however, no significant
correlation between maximum bicarbonate out-
put and maximum acid output (4). Nevertheless,
there was a highly significant correlation be-
tween maximum bicarbonate output and weight
of pancreas (4), a project continued by Jack Han-
sky and Oswaldo Tiscornia (5). Later, in London
with Gutierrez I was able to use our near maxi-
mal peak bicarbonate output after even purer
GIH secretin in humans with and without chronic
pancreatitis and compare peak bicarbonate out-

194

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

put with submaximal bicarbonate output to Boots
secretin.

Although in the dog, Perrier and I (4) found
no significant correlation between maximum bi-
carbonate and acid outputs, I was able with Gu-
tierrez to find a correlation in humans. Bicarbon-
ate secretory capacity in patients with DU was
normal and comparable to gastric acid secretory
capacity (7). It was the basal bicarbonate output
of the duodenal aspirate in patients with DU
which was only half that of control subjects and
which led to decreased neutralization of gastric
acid in the duodenum. However, I have rather
abandoned the pancreas in the last 20 years and
confined myself to the stomach.

I was also trained in David Dreiling's Exper-
imental Surgery Laboratory in Allan Kark's De-
partment of Surgery. Lewis Burrows, Walter
Wildstein, and I (8) studied the maximum hista-
mine response (MHR) of denervated fundic
pouches in dogs, and found the most repeatable
measurement was my PAO with a mean coeffi-
cient of variation of only 12%. We used this MHR
by PAO methodology to look at the effect of drugs
such as Mecholyl, secretin, and insulin. Mecholyl
could potentiate submaximal histamine acid out-
put to a PAO even greater than MHR (8).

Paradoxically there were increases of a third
to one half of MHR of Heidenhain pouches after
transthoracic vagotomy (9). However after
antrectomy there were marked decreases in
MHR; vagotomy after antrectomy now inhibited
instead of augmenting acid. We interpreted this
paradoxical acid stimulation by vagotomy as an
augmentation of gastrin release from diminished
fundal acid output, which had a bigger effect than
the decrease of gastrin release through antral de-
nervation. We could but speculate, because gas-
trin could not then be measured.

Our group also devised a protocol for measur-
ing, both before and after acid lowering opera-
tions, 12-hour overnight acid output, hourly pH,
and basal and peak acid outputs after an aug-
mented histamine test (10). This protocol was to
be used in a policy for selective surgery for DU,
but I left before this began.

The Spirit of Sinai. After four years at Ox-
ford, two years in the Army, and seven years at
the Middlesex Hospital in London, it was pro-
foundly stimulating to be in a different country, a
different city, and a different biomedical milieu.
Just as American cornflakes came in three sizes,
large, giant, and jumbo, so also did Sinai labora-
tory, clinical services, and staffing seem generous
beyond my experience, if hardly surprising, the
United States having twice the British GNP and

spending twice the percentage of GNP on health
than did the UK.

Sinai residents and fellows had qualified af-
ter eight years in college and medical school, com-
pared with our five or six, but then worked hard
and fast to become independent specialists within
five or six years, compared with our ten or twelve.
I spent two of my twelve months in the United
States visiting many of its great medical centers
and found some even more lavish, but none more
dynamic, than Sinai.

Sinai had a demonic pace, as each trainee
strived to acquire clinical and technical skills on
the one hand and then in research time to achieve
measurable success — "think it, do it, publish it."
Organizationally I have tried to emulate that
weekly timetable of gastrointestinal meetings
both alone and with radiologists, histopatholo-
gists, and surgeons. I still vividly recall sessions
with Franklin Hollander, Richard Marshak, Fen-
ton Schaffner, and Hans Popper, as well as Gut-
man's Medical Grand Rounds.

The only disappointments were a double dose
of parochialism. Nobody at Sinai asked me any-
thing about the British National Health Service.
When I returned to London nobody wanted to
hear anything about the health-care scene in the
United States.

Henry Janowitz. What were the secrets of
Henry Janowitz's achievements? First of course
his intellect, honed by generations of study. Sec-
ond, he had not only read everything, but remem-
bered only what was relevant. Third, agility: to
concentrate totally on the scientific or clinical
problem at hand, to comment positively (or neg-
atively) with suggestions of where to go next (or
to stop a project clearly going up a blind alley): he
could then put that problem out of his mind and
move onto the next resident or fellow. Fourth,
pragmatism: he had a shrewd idea of the assets
and deficits of those in his own and other units,
and he also knew how to live with them: "any big
unit should be able to tolerate at least one bas-
tard." Finally, conjugality, with the warm ambi-
ence and hospitality at his home with Addie — to
whom, and to Henry, in gratitude of both my year
at Sinai and our subsequent friendship, I wish not
Happy Retirement but Happy Continued Produc-
tivity.

References

1. Baron JH. Studies of basal and peak acid output with an

augmented histamine test. Gut 1963; 4:136—144.

2. Baron JH. An assessment of the augmented histamine

test in the diagnosis of peptic ulcer. Gut 1963; 4:243-
253.

3. Baron JH, Perrier CV, Janowitz HD, Dreiling DA. Max-

Vol. 60 No. 3

REMINISCENCES

195

imum alkaline (bicarbonate) output of the dog pan-
creas. Am J Physiol 1963; 204:251-256.

4. Perrier CV, Baron JH, Dreiling DA, Janowitz HD. Rela-

tionship between maximum bicarbonate and maximum
acid outputs in the dog. Proc Soc Exp Biol Med 1967;
124:312-314.

5. Hansky J, Tiscornia OM, Dreiling DA, Janowitz HD. Re-

lationship between maximal secretory output and
weight of the pancreas in the dog. Proc Soc Exp Biol
Med 1963; 114:654-656.

6. Gutierrez LV, Baron JH. A comparison of Boots and

G.I.H. secretin as stimuli of pancreatic secretion in hu-
mans with or without chronic pancreatitis. Gut 1972;
13:721-725.

7. Gutierrez LV, Baron JH. A comparison of basal and stim-

ulated gastric acid and duodenal bicarbonate in pa-
tients with and without duodenal ulcer disease. Am J
Gastroenterol 1976; 66:270-276.

8. Baron JH, Burrows L, Wildstein W, Kark AE, Dreiling

DA. The maximum histamine response of denervated
fundic pouches in dogs. Am J Gastroenterol 1965; 44:
333-345.

9. Baron JH, Burrows L, Wildstein W, Kark AE, Dreiling

DA. The effect of antrectomy and vagotomy on maxi-
mally stimulated canine Heiderhain pouches. Am J
Gastroenterol 1965; 44:467-475.
10. Baron JH, Burrows L, Wildstein W, Kark AE, Dreiling
DA. Unpublished data 1962. Cited in Baron JH. The
rationale of the different operations for peptic ulcer. In:
Cox AG, Alexander- Williams J, eds. Vagotomy on trial.
London: Heinemann, 1973, 6—35.

J. H. Baron, DM, FRCP, FRCS
Department of Surgery
Royal Postgraduate Medical School
Hammersmith Hospital, London

A tribute to Henry D. Janowitz from his most
distant metastasis: I joined the Division of Gas-
troenterology at The Mount Sinai Hospital in
July, 1962, as a research fellow. I had come for a
year's training, after spending a year at the Cen-
tral Middlesex Hospital in London with Francis
Avery Jones. Luminal gastroenterology at Mount
Sinai in those years was noted for focusing on two
areas: pancreatic pathophysiology and inflamma-
tory bowel disease. Henry Janowitz decided that I
was made for the pancreas, and I spent the ensu-
ing 12 months alternating between dogs equipped
with Thomas cannulae in the Atran laboratory
and intubating patients with Dreiling tubes for
secretin tests. Henry Janowitz and David Dreil-
ing took me under their wings and stimulated my
first love in gastroenterology — the pancreas. The
year was very productive in terms of publications,
research work done, and techniques learnt.

What were some of the memories from that
year? Tutorials with and presentations to Frank
Hollander on gastric and pancreatic physiology;
Friday lunch meetings with David Dreiling and
the dog team, where I was introduced to pastrami

on rye and Dr. Pepper's; Grand Rounds under the
stern chairmanship of Alex Gutman; learning
about serotonin from Dick Warner and the esoph-
agus from Bernard Cohen; doing rounds with
Henry Janowitz, Nat Cohen, Larry Werther, Jeff
Friedman, Al Gelb, Art Lindner; and sharing
knowledge with Jesse Berkowitz, Richie Rosen-
berg, Jared Kniffen, Jerome Waye, Jacques
Kessler, Hugh Baron, Oswaldo Tiscornia, Mort
Davidson, and Gene Aronow. I also recall my ner-
vous presentation of some aspects of pancreatic
physiology at the Federation meeting in Atlantic
City where Mort Grossman made incisive com-
ments that have lasted for 30 years. I particularly
treasure memories of the fantastic hospitality
shown by all at Mount Sinai, but especially by
Eileen and Nat Cohen, Addie and Henry Janow-
itz, and Muriel and David Dreiling, who made life
in Manhattan very exciting for Paula and me and
our two small children.

If one examines the names of the people who
were at Mount Sinai in 1962-1963, one finds that
they have all made significant contributions to
gastroenterology over the years and have paved
the way to make the Division of Gastroenterology
at Mount Sinai one of the best in the United
States.

Three people most influenced my career in
gastroenterology — Sir Francis Avery Jones,
Henry Janowitz, and Mort Grossman. Both
Grossman and Janowitz worked with Ivy in Chi-
cago, and both were considered brilliant gastro-
intestinal physiologists, but they then proceeded
on divergent paths — Grossman to a full-time
physiologic research career and Janowitz to a
mixture of patient care and research.

Henry Janowitz did both activities exception-
ally well. He molded a strong department with
excellent clinical ramifications together with a
research team that produced world-class research
and publications in "good" journals. His program
attracted applicants from all of the United States,
and the excellence of his program also influenced
a number of foreigners to apply to work with him
at Mount Sinai. Thirty years ago, the number of
training posts in the country was not as vast as it
is now, but one of the best was at Mount Sinai. I
was fortunate that Henry had a position available
when I was ready to come to the United States
and that there was such a vast array of clinical
material to learn from. The year proved very
fruitful in terms of research papers published —
there were seven or eight — and techniques mas-
tered. It was, of course, not only luminal gastro-
enterology that was so well taught, but also liver
disease with Fenton Schaffner and Hans Popper.

196

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

My year at Mount Sinai with Henry Janowitz
was of such moment that it influenced my deci-
sion to take up a full-time academic post in Mel-
bourne, Australia, where I could try to emulate
Henry — there was a goodly blend of research,
teaching, and patient care. On my return to Aus-
tralia, I tried to continue my research in basic
pancreatic physiology but was unsuccessful with
Thomas cannula dogs, so my energies went in-
stead into clinical pancreatic function tests, with
development of a maximal test for pancreatic se-
cretion; into esophageal motility; and into radio-
immunoassay of the hormone gastrin. My train-
ing at Mount Sinai under the expert tutelage of
Henry D. Janowitz was the key to my success as a
gastroenterologist, researcher, and teacher, and I
am very proud when Henry Janowitz refers to me
as his "foremost metastasis."

I could conceive of no greater honor than to
have a unit named after oneself (whilst still liv-
ing and active). Although Henry has had many
honors bestowed upon him, I am certain he will
cherish the present one the most. Henry D. Jan-
owitz— well deserved!

Jack Hansky, M.D.
Assoc. Prof., Monash Medical Center
Clayton, Victoria, Australia

In 1968 I had an interview with Henry Janowitz
as a requisite to becoming a gastrointestinal fel-
low. He accepted me and I had to start in July
1969. However, I could not start on time because
I was finishing the gastrointestinal residency in
Venezuela. I wrote a letter to Henry and he an-
swered, "So if you start a few weeks later, you will
win the Nobel Prize a few weeks later." It was my
first experience with Henry's sense of humor.

Coming to New York with my wife and two
babies was not easy, but Henry Janowitz made
sure that I felt at home beginning with my first
week in his department. I was treated with
friendliness and respect, which at that time was a
much-needed sensation.

My two years as a gastrointestinal fellow at
Mount Sinai (1969-1971) were exciting and pro-
ductive. I found a group of hard-working people
doing the most varied research, combined with
excellent work under Henry's supervision.

After the first month I started to work at the
laboratory on the effect of ATPase inhibitors and
cyclic AMP on pancreatic secretion. For me, this
experience was the most rewarding. I learned
how to do research. I remember meeting with

Henry at 9 or 10 pm to discuss the results of my
work. I had never before had a meeting to discuss
ongoing research that late at night. However,
Henry was different. He was an extremely hard-
working man. He has been a model for me.

At medical rounds Henry was a master. Be-
ing an excellent clinician, and a smart one, he
allowed and encouraged everybody, including the
fellows and young physicians, to give their opin-
ions before he gave his own. Sometimes, during a
clinical discussion of a difficult case, one of the
young residents — David Sachar — left the room to
look for some unusual information on the case in
his "computerized" hole-punch card system. I en-
joyed these discussions and today, I manage clin-
ical discussions in a style I learned from Henry. I
learned endoscopy at Mount Sinai, and I was
lucky to have Jerome Waye also learning how to
use the Gastrocamera and the new fiberoptic
scopes. Also, I had the opportunity to work in the
motility laboratory with Bernie Cohen, who
taught me esophageal motility and even how to
prepare my own esophageal catheter. Since then,
I have been a "motility man."

Today, as chief of a gastrointestinal depart-
ment in a teaching hospital in Caracas, Venezu-
ela, I am sure that my directing style has been
influenced by my experience in Henry Janowitz's
department.

Moises Guelrud, M.D.
Chief, Gastroenterology Department
Policlinica Metropolitana
Caracas, Venezuela

As part of my gastroenterology training at the
Mount Sinai Medical Center, I was resident to
Dr. Henry Janowitz for six months in 1972, I

have always felt that it was an honor and a priv-
ilege to have held this position. Through his en-
deavors as a clinician, scientist, and humanist.
Dr. Janowitz has had a profound influence on my
career.

The most important impact has been in the
area of time management. As is the case with
most academic physicians, my activities include
patient care, teaching, research, and administra-
tion, two or more somehow frequently occurring
simultaneously. Working with Dr. Janowitz
taught me four "rules" that govern most of my
time. Two apply to all aspects of my work, and the
other two concern clinical practice.

Henry's first rule: Focus. Do only one thing at
a time, and do it well. This doesn't mean you must

Vol. 60 No. 3

REMINISCENCES

197

finish one item before beginning another; it sim-
ply means don't do two things at the same mo-
ment.

His second rule: Be flexible and adaptable. A
vignette comes to mind to illustrate this point.
One day, we were asked to see a man with
Crohn's disease and bacterial overgrowth second-
ary to an ileal stricture. We were told that the
patient was rather compulsive and had requested
that only Dr. Janowitz should question him and
examine him. When we entered the room, the
man was washing his hands. In the course of get-
ting ready to get into bed for the physical, he re-
turned to the sink several times to wash and re-
wash his hands. As Dr. Janowitz was about to
begin the examination of the abdomen, the pa-
tient said, "Just a minute. You can only examine
my abdomen if you are prepared to follow certain
rules! You can start wherever you want, but once
you touch me below the belly-button, you are not
allowed to touch me above the belly-button
again!" This led to a rather unorthodox method of
abdominal examination; nevertheless, it was ac-
complished without discussion or complaint.

Henry's third rule: Take your time. He al-
ways took as long as was necessary to get the
complete story, so that the correct decision would
be made.

His fourth rule: If your patient needs a sur-
geon doctor, he or she still needs a doctor doctor.
This was not meant to say that the surgeon is
merely a technician, only that most gastrointes-
tinal problems are codisciplinary. As Dr. Janow-
itz's resident, I had to report to him from the op-
erating room to let him know if the findings were
as expected. If not, then I was the go-between for
the discussion, subsequent to which a joint deci-
sion was made. I have followed this practice for
the past 21 years. My surgeons have come to ac-
cept and appreciate my input.

My flrst 14 years were almost entirely clini-
cal. In the past seven years. Dr. Janowitz's thirst
for knowledge through research has had an in-
creasing influence on me. Since becoming head of
our division, I have developed an active clinical
trials program and have become the codirector of
a research laboratory.

In summary, Henry Janowitz has had a per-
vasive effect on all aspects of my career. I feel that
I have followed in his footsteps, though my shoe
size will always remain smaller than his.

Fred Saibil, MD, FRCPC
Head, Division of Gastroenterology
Sunnybrook Health Science Center
Toronto, Ontario, Canada

We surgeons have known, worked with, been
taught by, and admired Henry for almost 40
years. My colleagues and I never cease to be
amazed by him. He is a superb physician whose
clinical acumen and judgment are brilliant. To
us, Henry represents the ultimate gastroenterol-
ogist and gastroenterological consultant. It is a
rare privilege indeed to have been asked by Dr.
David Sachar, chief of the Dr. Henry D. Janowitz
Division of Gastroenterology, to represent the De-
partment of Surgery in this tribute to Henry D.
Janowitz.

Henry is a man of few words, but always lis-
ten carefully to him; you will learn much. He is
pragmatic but rarely dogmatic. If you choose to
disagree or argue an opposing view, be sure of
your facts, because Henry's experience and
knowledge are so vast that you will rarely win.
By the same token, when you are confronted with
a decision on whether or not to operate, Henry's
opinion can be all-important. He has the amazing
ability to know the therapeutic limitations of
both medical and surgical management in any
given case.

Henry has published extensively, but to me
his book Inflammatory Bowel Disease — A Per-
sonal View, above all others, contains a treasure
trove of helpful information in caring for the pa-
tient with IBD. Although Henry is rarely dog-
matic, he entitled Chapter 3 "The Nature of the
Beast: A Speculative Chapter with Some Dog-
matic Comments." But then, in typical Henry
Janowitz fashion, the first sentence of the chapter
notes that "readers who find speculation unprof-
itable may want to skip this Chapter and move
on." There are multiple quotable quotes in this
book. In a discussion of the rarity of appendicitis
in Crohn's disease and the occurrence of isolated
Crohn's disease of the appendix, Henry notes,
"[My] impression must have some validity." He is
absolutely on the money when he states: "most
medical errors, I believe, are errors of omission.
What we do is essentially rational; it is what we
do not do that is important." This could well rep-
resent the surgeon's as well as the internist's
credo.

Those of us who have the current good for-
tune to be members of the Department of Surgery
at Mount Sinai stand on the shoulders of giants.
We recognize the enormous debt we owe to
Drs. Garlock, Ginzburg, Colp, Klein, Dreiling,
Lesnick, Kirschner, Klingenstein, and others for
their teachings. Our department is recognized as
a major center for gastrointestinal surgery. In
having achieved this designation, we are ever

198

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

mindful of and pay tribute to Henry and his dis-
ciples for the role that they have played in our
success. Not only has Henry been personally in-
volved in the care of so many of our patients, but
he has also been a major contributor to our re-
search endeavors and teaching activities. Henry
has been the mentor of almost all of the gastro-
enterologists who now make up the superb Dr.
Henry D. Janowitz Division of Gastroenterology,
so ably headed by Dr. David Sachar. They have
formed a liaison with us in a spirit of collegiality
that is the envy of many academic health centers.
Together we have been able to advance medical
knowledge, render superb patient care, and pro-
vide the finest teaching in the field of gastroin-
testinal disease.

Henry — we salute you and congratulate you
on the occasion of this Festschrift in your honor.

Arthur H. Aufses, Jr., MD
Franz W. Sichei Prof, of Surgery
Mount Sinai School of Medicine
Director, Dept. of Surgery
The Mount Sinai Hospital
New York, NY

Twenty -five years ago, Henry D. Janowitz, M.D.,

listened to the frustration of the families of those
who suffered from Crohn's disease and ulcerative
colitis and gave them the precious gift of hope. He
challenged them to raise funds to develop the first
national research program dedicated to ending
the scourge of these inflammatory bowel diseases
(IBD). In exchange, he promised to lend his pres-
tige to their undertaking and helped establish the
Crohn's & Colitis Foundation of America (CCFA).
He provided guidance and leadership to this
fledgling organization, fostering the growth of
IBD research across the United States. Indeed, as
much as any single physician, Dr. Janowitz has
had the most dramatic influence on research that
seeks to find the cause of, and cure for, IBD.

From 1967 to 1975, Dr. Janowitz served as
the National Scientific Advisory Chairman for
CCFA. During his tenure, he nourished the
dreams of scientists and lay persons alike. Pains-
takingly, he expanded CCFA's research program.
From initial projects that sought to define more
clearly the clinical aspects of IBD and to establish
a basic epidemiological profile of patients,
CCFA's research program expanded its scope to
include basic research that began to investigate
the role of the immune system in IBD and to de-

termine whether or not infectious agents trig-
gered its onset. Reaching out beyond the New
York community of researchers, he encouraged
scientists at such prestigious institutions as the
Johns Hopkins University School of Medicine, the
Cleveland Clinic, the University of Chicago, Mas-
sachusetts General Hospital, and Cedars-Sinai
Medical Center in Los Angeles to focus their stud-
ies on IBD. By 1975, CCFA had funded research
grants at 21 hospitals in 13 cities, totaling
$1,470,433.

When Dr. Janowitz stepped down as the
chairman of CCFA's National Scientific Advisory
Committee, CCFA was firmly established as a re-
spected presence in the research community. As a
testament to his work, one of America's leading
clinicians and researchers in the field of IBD ac-
cepted the chairmanship of CCFA's National Sci-
entific Advisory Committee. He was Joseph B.
Kirsner, M.D., Ph.D., Louis Block Distinguished
Service Professor of Medicine at the University of
Chicago.

Because of Dr. Janowitz's belief in the ability
of research to conquer IBD, his extraordinary tal-
ent as a mentor, and his enormous enthusiasm,
CCFA is now one of the premier health organiza-
tions in America. Since 1967, CCFA has invested
more than $19 million in research projects in the
United States, Europe, and the Middle East. Its
researchers have identified a mouse model, are
building a cell-line bank, and are close to identi-
fying the gene markers for Crohn's disease and
ulcerative colitis. They also are making exciting
headway in their understanding of the role of the
immune system. In addition, they have improved
medical and surgical treatment of these diseases
and sharpened pathologists' diagnostic skills.
Their success has engendered greater interest in,
and support of, IBD research at the National In-
stitutes of Health. For each of the past three
years, the federal government has earmarked $2
million specifically for IBD research.

Hundreds of thousands of people who have
Crohn's disease or ulcerative colitis have received
educational material and supportive services pro-
vided by CCFA. Once a "closet" disease, IBD is
now regularly discussed in major newspapers and
national magazines and on radio and television
talk shows. Celebrities and athletes throughout
the United States speak out in behalf of CCFA's
research and education programs.

In the mid-sixties. Dr. Janowitz fostered the
dream of Irwin Rosenthal, an attorney, and
William Modell, a businessman. Together, these
three men gave birth to a national organization of

Vol. 60 No. 3

REMINISCENCES

199

dedicated medical and lay volunteers who one day
will find a cure for inflammatory bowel diseases.

On behalf of the Crohn's & Colitis Founda-
tion of America and the estimated two million
Americans who suffer from the inflammatory

bowel diseases, I proudly salute and honor a great
physician, researcher, and humanitarian.

Jane W. Present
Chairman of the Hoard
Crohn's & Colitis Foundation of America, Inc.

Grand
Rounds

Obscure Gastrointestinal Bleeding

Blair Lewis, M.D.

I WANT to review here the experience at the
Mount Sinai Medical Center with gastrointesti-
nal bleeding of obscure origin and explore how
many different doors of clinical research have
been opened by that experience.

A simple case presentation exemplifies the
type of problem I am addressing. A 73-year-old
man I saw recently had been bleeding for the last
two years. During those two years he received 30
units of transfused blood. He had two small-bowel
series, two barium enemas, four upper en-
doscopies, five colonoscopies, a couple of bleeding
scans, an angiogram. He continues to bleed from
an unknown site.

Fortunately, the site of bleeding is identified
in most patients who have gastrointestinal bleed-
ing. Only an estimated S%-5% of patients who
bleed from a gastrointestinal source have obscure
gastrointestinal bleeding — bleeding whose site
cannot be found despite routine testing.

Several reasons account for the inability to
identify the site of bleeding:

• The bleeding rate may be so slow as to make
certain tests, such as bleeding scan or angio-
gram, unsuccessful at showing extravasation of
blood.

• Bleeding can be intermittent; the person may
be bleeding in the physician's office, but when
that patient gets to the emergency room, he or
she has stopped bleeding, and whatever testing
for extravasation of blood is performed from
then on will be unsuccessful as well.

• The majority of these lesions are vascular ecta-

Adapted from the author's Grand Rounds in Medicine presen-
tation at the Mount Sinai Medical Center on March 19, 1991.
Final revision received October 1991. From the Department of
Medicine, Mount Sinai Medical Center, New York. Address
reprint requests to the author, who is an Assistant Clinical
Professor of Medicine at Mount Sinai School of Medicine, at
1067 Fifth Avenue, New York, NY 10028.

200

sias. Endoscopists have learned that these le-
sions can vasoconstrict right before the physi-
cian's eyes, and disappear; a physician may be
looking at one one moment and not see it the
next. That makes diagnosis difficult.
• Even if a bleeding site is found and treated, by
whatever method, the patient may go on to
bleed, because the true cause of bleeding has
not been discovered. Not everything the physi-
cian finds and treats will be the actual cause of
blood loss.

Causes

Angiodysplasia. Angiodysplasias have many
different names: arteriovenous malformations,
vascular ectasias, angiomas, arteriovenous fistu-
las. Since 1977, when Scott Boley first taught us
about these vascular ectasias, our knowledge has
increased. Boley originally said that angiodyspla-
sia was a disease of the elderly (patients were
over the age of 60); that the lesions most com-
monly occur in the cecum or the right colon; that
the lesions were small (1). He said that they were
diagnosed by angiography and not by endoscopy,
which we now know is not true; endoscopy does
play a major role in diagnosis. Yet the true pa-
thology or pathophysiology of angiodysplasia is
still unknown.

We now know certain other factors about an-
giodysplasias. In addition to a predilection for
the elderly, the disease tends to be associated
with certain illnesses. Aortic stenosis has been
thought to be a factor in the development of an-
giodysplasia. Recent literature and prospective
studies using echocardiography along with endos-
copy and angiography have shown that aortic ste-
nosis is not a cause of angiodysplasias and is not
a risk factor independent of age (2). We know that
renal failure is associated with the development
of vascular ectasias, usually in the stomach. This
association was established by Zuckerman, who

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

OBSCURE GI BLEEDING— LEWIS

201

showed that when patients with acute upper gas-
trointestinal bleeding were examined by endo-
scope, vascular ectasias were found in the major-
ity of those with renal failure, but in only a
minority of those without renal failure (3). Other
diseases, such as Osler-Weber and von Wille-
brand's disease, are widely known to be associ-
ated with telangiectasias.

How do these angiodysplasias form? Scott Bo-
ley formulated a theory about the development of
angiodysplasia that is widely accepted, but still
debated; it remains a theory (1). Some knowledge
of microcirculation in the colon is needed to talk
about his theory (Fig. 1).

Boley hypothesized that distension of the
cecum and the ascending colon leads to chronic
intermittent obstruction to venous outflow. If this
process continues over many years, the capillary
beds (honeycomb complexes) become dilated,
whereupon the back pressure could then be fed to
the arteriolar side. Finally, the arteriolar side
would lose its prearteriolar sphincter, causing in
essence an arteriovenous fistula and further dila-
tion of the capillary beds and forming an angio-
dysplasia, which could then bleed.

Other Causes. Angiodysplasia is not the only
cause of obscure bleeding. Among the many other
causes are small-bowel tumors and ulcerations of
the small intestine (Table).

Nonenteroscopic
Diagnostic Tests

Small-Bowel Series. A variety of tests are
available to help make the diagnosis in these pa-
tients. But I must adjure physicians not to forget
the basics. I think that we endoscopists have been
very good at brainwashing the medical commu-
nity into thinking that endoscopy is the end an-
swer, and to forget about the basic tests, upper
gastrointestinal series and barium enema exam-
inations. For example: a 67-year-old man came to
The Mount Sinai Hospital with melena. No ab-
normalities were detected in an upper intestinal
endoscopy or colonoscopy. The treating physi-
cians said, "Blair, why don't you do an enteros-
copy to try to figure out where the bleeding is
coming from?"

I responded, "Why don't you do a small-bowel
series?"

"Well," they said, "the yield of small-bowel
series is low."

I said, "Yes, but it's easy to do, it's inexpen-
sive, it's not traumatic for the patient, it's not a
lengthy examination." Moreover, it was part of
our protocol for doing enteroscopy.

Fig. 1. Colonic crypts are fed by arterioles that come up from
the serosal layer and then form circles at the tops of the crypts.
These circular capillary beds are called honeycomb complexes.
Venules drain down through the muscularis propria and into
the veins. Reproduced with permission from ref. 1.

So they did a small-bowel series and within
the third portion of the duodenum (Fig. 2) was an
annular lesion, a pancreatic cancer that was
growing into the duodenum. That was the cause
of bleeding in this patient.

The small bowel has been considered rela-
tively inaccessible in the past, and the develop-
ment of enteroscopy does allow direct visual in-
spection of the intestine. Routine upper intestinal
endoscopy does not reach the third part of the
duodenum; it will reach the second portion, and
occasionally the junction of the second and third
portions. This case, in which the diagnosis was
not, and could not have been, made by routine
upper endoscopy, shows how helpful a small-
bowel series can be, even though the yield is low.
In a recent series of a hundred patients with ob-
scure bleeding, no diagnoses were made with the
small-bowel series (4). Still, a small-bowel series
should be considered one of the basic diagnostic
tests.

TABLE

Causes of Obscure Bleeding

Brisk bleeding Occult bleeding

Angiodysplasia

Leiomyoma

Adenocarcinoma

Leiomyosarcoma

Lymphoma

Jejunal diverticula

Carcinoid

Crohn's disease

Aortoenteric fistula

Zollinger-EUison

Meckel's diverticulum

Vasculitis

Duplication cyst

Potassium

Hemangioma

Infectious causes

Ulcerative jejunitis

Fig. 2. Small-bowel series revealing cause of bleeding: an-
nular lesion in third portion of duodenum, which routine up-
per intestinal endoscopy cannot reveal.

Fig. 3. Scan shows blood within jejunum. Sometimes nuclear
medicine physicians find it hard to tell what piece of bowel
this is in. They need to follow the scan over time to see the
serpiginous route and the contractions of the small intestine,
versus the fixed locations of the colon, to help differentiate
those two.

Fig. 4. Enteroclysis of a 46-year-old man who had obscure
bleeding; a leiomyoma sits in wall of jejunum.

Fig. 5. Cecal angiodysplasia. The two cecal arteries take
parallel courses. Some tufting within base of cecum, and an
early-filling vein up the middle, characteristic of vascular
ectasia. Courtesy of John Train.

Vol. 60 No. 3

OBSCURE GI BLEEDING— LEWIS

203

Scans. How about scans? The days of sulfur
colloid scans — those very short scans — are gone.
We used to inject sulfur colloid; if it extravasated,
a quick blush would appear, and if the person was
bleeding, within the three-minute duration of the
examination, the physician would be able to pick
it out. Nowadays, we withdraw blood from a pa-
tient, label it with technetium 99m, and reinject
the blood. Patients can then be followed up for as
long as 24 hours. As little as 5 cc of intraluminal
blood will show up as a blush on the scan screen
(Fig. 3).

Still, scans for bleeding do not delineate the
exact cause of bleeding within the small intes-
tine: they indicate only the presence and the lo-
cation of blood. Unfortunately the yields are ex-
tremely low with scans.

Enteroclysis. Enteroclysis is not available in
many places; in all of Manhattan, I think only one
center offers it. Enteroclysis is a double-contrast
study of the small intestine. A tube is placed
through the patient's mouth or nose and ad-
vanced through the stomach to the duodenum.
Barium is injected through the tube and fills the
small intestine without filling the stomach,
avoiding any overlay of stomach and small intes-
tine; methylcellulose can be given as a second
contrast agent to give beautiful films (Fig. 4).

It has been reported in the literature that
enteroclysis is a gold standard for small-bowel tu-
mors. A positive test indicates a small-bowel tu-
mor; no one doubts that. Unfortunately a nega-
tive study does not effectively exclude the
presence of a small-bowel tumor; for example, the
yield of enteroclysis in a recent large series in
patients with obscure bleeding was 10% (5).

Angiography. The medical community puts a
great deal of faith in angiography, and it is cer-
tainly one of the standard ways to make a diag-
nosis. Angiography for angiodysplasia is done in
several different ways. Certain signs are charac-
teristic: a slowly emptying vein, an early filling
vein, and a vascular tuft (Fig. 5). This work again
came from Boley's group at Montefiore (6). Ex-
travasation in Boley's series was quite rare.

But are these signs clinically useful? In 1989,
the Montefiore group reported that within the
small intestine the yield of angiography in ob-
scure bleeding was 14% (7). Angiographic diagno-
sis of angiodysplasia in the small intestine is dif-
ficult, one of the reasons being that there are so
many vascular arcades within the small intes-
tine.

Why not heparinize the patient and then per-
form therapeutic angiography? This has been rec-
ommended in the past. Some time ago I received a

pamphlet in the mail about the diagnosis of ob-
scure bleeding and its management. I picked up
the phone and called the 1-800 number for the
promised lecture, which instructed me that if a
patient has a negative endoscopy and a negative
angiography, I should give the patient anticoag-
ulation agents and then re-angiogram the pa-
tient. That course of action has been life-threat-
ening in every patient referred to me after
treatment with stress aspirin doses, heparin, or a
variety of anticoagulation agents. I have never
seen that treatment to be useful; it is risky and I
cannot advocate it.

Diagnostic Surgery. If all these simple
straightforward tests do not confirm where the
bleeding is coming from, why not just operate on
the patient, because the disorder is not really a
medical but a surgical illness? However, surgery
without some type of guidance almost always
ends in failure. The literature on the yield of sim-
ple exploration dates back to the preendoscopic
era, to 1961, when a group out of Boston reported
on a hundred patients with obscure bleeding who
were taken to the operating room (8). This was
before the days of endoscopy, so that they were
making the diagnosis at surgery of esophageal
varices, or peptic ulceration, or colon cancer, dis-
orders now diagnosed preoperatively. Removing
those diagnoses from the results and looking only
at the yield for disorders we think routine en-
doscopies and barium studies would have missed,
the yield is 13%. So simple exploration is some-
thing that no one should be prepared to do in the
obscure bleeder. It has to be coupled with some
other form of identification. That is why Jerome
Waye of this institution turned to enteroscopy, to
try to help these patients; I was fortunate enough
as a fellow to be introduced to enteroscopy 6 years
ago.

Types of Enteroscopy

The small bowel can be examined endoscopi-
cally in several ways:

• Routine upper intestinal endoscopy, esophago-
gastroduodenoscopy (EGD), can reach the junc-
ture of the second and third portions of the du-
odenum.

• Push-enteroscopy examination, also called "ex-
tended EGD," uses a longer ifistrument — some-
times a colonoscope — to reach two feet beyond
the ligament of Treitz into the jejunum.

• Small-bowel enteroscopy, also called sonde-en-
teroscopy ("sonde" meaning balloon in French),
utilizes a 9-foot-long endoscope with a balloon
at its tip that allows peristalsis to carry the

204

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

instrument down through the gut. The small
intestine is then examined during the instru-
ment's removal. Nearly the whole small intes-
tine can be examined in this fashion.

All of these procedures can be performed intraop-
eratively.

Push enteroscopy is straightforward and is
routinely performed. It can be used to make diag-
noses of patients with obscure bleeding, as well as
to obtain small-bowel biopsies in patients with
malabsorption, or place jejunal feeding tubes.
These instruments can be used both diagnosti-
cally and therapeutically; in the duodenum it is
possible to see a vascular ectasia and then to
place a coagulator through the instrument and
cauterize the lesion.

But push enteroscopy still leaves 8-10 feet of
small intestine unexamined, and small-bowel en-
teroscopy, sonde-enteroscopy, has been developed
to try to examine that portion of the bowel. Other
instruments have been developed, but the state-
of-the-art instrument now used in several centers
around the country is the 9-ft instrument with its

Fig. 6. X-ray showing total small-bowel intubation with en-
teroscope. Tip of instrument sits at ileocecal valve in right
lower quadrant.

tip balloon. It is a thin instrument, 5 mm in di-
ameter, which is easily and comfortably passed
through the patient's nostril. It is not therapeutic,
it is purely diagnostic; the bowel wall cannot be
marked with it.

The endoscopic portion of the examination is
performed during withdrawal, pulling the instru-
ment back, looking through the instrument at
that point (Fig. 6).

Push enteroscopy and small-bowel enteros-
copy are intimately related, because the small-
bowel instrument is placed with a push instru-
ment. The small-bowel enteroscope does not
permit an adequate examination of the trans-
verse duodenum and the proximal jejunum. Ini-
tially a push enteroscope is placed, and the first
two feet of the jejunum are examined. If a bleed-
ing site is found there, the rest of the examination
is not necessary. The push instrument can then
be brought back to the stomach to grasp the little
transnasal small-bowel enteroscope, carry it into
the jejunum, and leave it there. That really short-
ens the examination time: previous experience in
Japan allowed passive passage through the py-
loris, in which patients lay on their right side and
waited for the instrument to pass in examinations
that extended over 24 hours and more. Jerry
Waye's idea of placing it beyond the ligament of
Treitz at the start of the procedure means that we
can do the entire examination in six to eight
hours.

I have now performed more than 480 entero-
scopic examinations (9). All patients had obscure
bleeding lasting on average over 2 years. Each
patient had been transfused an average of almost
18 units of blood and had received over 2 units a
month because of ongoing bleeding. All patients
had extensive evaluations, more than 1300 upper
intestinal endoscopies and 1200 colonoscopies.
Many patients had enteroclysis studies and angi-
ography that were nondiagnostic. The average
number of examinations for each patient prior to
enteroscopy was about 10. We made a diagnosis
in approximately 40% of those patients. Vascular
ectasias made up the majority of findings, ac-
counting for almost 80%. Small-bowel tumors
were the second most common diagnosis (dis-
cussed below). We have also found a variety of
other lesions, such as Meckel's diverticulum, di-
verticulosis of the small intestine, and aortoen-
teric fistula. The yield of push enteroscopy alone
is approximately 19%. The additional yield by
performing sonde-enteroscopy is 29%. There is
some overlap between those two groups; for ex-
ample, a vascular ectasia is found on the push
enteroscopy, then deep in the ileum another le-

Vol. 60 No. 3 OBSCURE GI B

sion is revealed on sonde-enteroscopy. Thus the
combined yield is not simply additive; the true
yield overall runs about 40%.

Differentiating Small-Bowel Tumors from
Vascular Ectasias. Mount Sinai Hospital has
some special experience with small-bowel tumors.
Asher Kornbluth and I worked together looking
at a group of patients who had small-bowel tu-
mors diagnosed at enteroscopy (10). Among the
first patients who had small-bowel tumors was
one who had an adenocarcinoma of the distal je-
junum (Fig. 7) that enteroclysis failed to identify.

Asher tried to see if there is a way to differ-
entiate patients with small-bowel tumors from
patients with vascular ectasias. You might think
that people with small-bowel tumors would have
some symptoms: maybe a little weight loss, or a
little vomiting, or more copious or more episodic
bleeding than patients with vascular ectasias. It
was thought that there must be some factor in the
clinical history to differentiate those two groups
of patients. In fact there is not. When Asher
looked at these two groups, they bled similarly, in
similar amounts and episodes, of maroon blood or
melena, unlike people who truly bled occultly
with just fecal occult blood testing. None of these
patients had any symptoms whatsoever.

The only real difference was age. It was con-
cluded that patients who were younger have a
greater risk of having a tumor as a cause of ob-
scure bleeding than a patient over the age of 60.

After the Diagnosis

What do we do after we make the diagnosis?
Surgery seems to be the most straightforward.
For a vascular ectasia deep in the small intestine,
why not just cut it out? That certainly is a viable
approach. But the point is that the person cannot
be sent to the operating room simply because
somewhere in the jejunum is a vascular ectasia. It
has to be localized. The small intestine is differ-
ent from the colon. In the colon, the surgeon can
proceed to take out the right portion or the left
portion, as necessary; resection of either portion
is a viable plan. But large areas of the small
intestine cannot be resected; the person is at risk
of becoming a nutritional cripple with terrible di-
arrhea. The site of bleeding must be marked to
limit the resected area.

Certain intraoperative tests have been advo-
cated in the literature to try to help with this
problem. Research on two different approaches,
intraoperative scintigraphy and intraoperative
endoscopy, has been conducted at Mount Sinai.

Intraoperative Scintigraphy. Intraoperative

LEWIS 205

Fig. 7. Adenocarcinoma of large circumference, nearly ob-
structing the intestinal lumen, that enteroclysis failed to iden-
tify. Enteroclysis is not the gold standard many radiologists
believe it to be; enteroclysis of this patient revealed no abnor-
malities.

scintigraphy was actually developed at The
Mount Sinai Hospital. Christopher Palestro, a
specialist in nuclear medicine, has published ex-
tensively on it (11). It is a fantastic technique that
will, it is to be hoped, catch on in the surgical
world. What Pallestro suggested was that if a pa-
tient has a bleeding scan indicative of bleeding in
the small intestine, they be whisked to the oper-
ating room right away, bringing the nuclear med-
icine scanner along. Then bring out the bowel and
scan it to see which portion of bowel contains the
radioactive tracer, the radioactive blood. Much
run-off takes place when the bowel is manipu-
lated, so the first step in the procedure is to clamp
the bowel every 30 cm or so. The blood does not
move off somewhere else and give a false reading.
Then each bit of clamped bowel is brought under
the scanner and examined for the one segment
that "lights up." We have recently published a
report on a series of patients managed in this
way. For this approach to work, the patient must
have a positive bleeding scan or a positive angio-
gram— must, in other words, be actually bleed-
ing.

Intraoperative Endoscopy. Intraoperative
endoscopy does not require that the patient be
bleeding at the time of surgery. It is possible to
put an endoscope into an anesthetized patient and
pleat the entire small intestine over the endo-
scope looking for bleeding sites (Fig, 8). At any
bleeding site, the surgeon puts a small suture
through the serosa marking the spot; then the

206

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

Fig. 8. Angiodysplasia within small bowel using intraoper-
ative endoscopy. An endoscopist knows instantly this is an
intraoperative photograph, because room lights are visible
through the wall of the bowel — a sign at a regular endoscopy
that something has gone wrong.

examination continues throughout the small in-
testine, each bleeding site being marked with a
suture. Finally, the surgeon reviews the sutures:
"Three marks here in this part of the bowel, one
mark here ..." and can decide what type of re-
section, what type of margins, are preferable.

We have recently reported on a series of 23
patients with obscure bleeding who went to sur-
gery using intraoperative endoscopy (12). Given
the extensive literature about the yield of intra-
operative enteroscopy, we wanted to see if preop-
erative enteroscopy was a help, a hindrance, or
neutral in the management of these patients. The
23 patients were similar to the group discussed
above; the average age was 72 years, they had
been bleeding for two years, had received an av-
erage 28 units of blood, and generally had waited
about three months before they decided to go
ahead, or their physicians decided to go ahead,
with surgery.

Exact correlation between small-bowel en-
teroscopy and intraoperative endoscopy was ob-
tained in more than three quarters of the pa-
tients. But there were discrepancies between
those two examinations. The transnasal scope
missed lesions in two patients. The instrument
does not have a tip to flex, unlike a regular endo-
scope. There are areas of bowel that are not ex-
amined. We estimate between 50% and 70% of the
bowel wall is examined with this procedure. In-
terestingly, intraoperative endoscopy also had a
miss rate, and so we considered these two exam-

inations complementary to each other. Using
both the preoperative and the intraoperative ex-
amination, the yield was better.

Unfortunately, up to 50% of the patients who
had surgery in a two-year follow-up bled again.
Clearly, surgery is not the final answer for these
patients. There has been interest in medical ther-
apy using hormonal agents, which has come to
the fore within the last couple of years. I think
probably it came to the medical community's at-
tention when Bronner reported on 7 patients who
had bleeding from vascular ectasias (13). They
were hemodialysis patients, and he treated them
with a combination of estrogen and progesterone.
Four of the 7 patients stopped bleeding during the
follow-up period. So people thought hormonal
agents would be perfect for people bleeding from
vascular ectasias. There was no control group in
his study, but Bronner used his own patients ret-
rospectively as their own control group. They
were bleeding before they were put on therapy;
they stopped on therapy; and that was considered
an excellent result.

Last year, van Cutsem reported a crossover
trial using hormonal agents in people with ob-
scure bleeding (14). Half of his patients had Os-
ier-Weber-Rendu or von Willebrand's disease, so
it was not an idiopathic group; these were people
who had associated illnesses. Still, patients
stopped bleeding when the therapy was started
and bled again when the therapy was discontin-
ued. We thought, "This is great," because we have
patients with diffuse ectasias of the small intes-
tine who are elderly, who cannot undergo surgery
or risk a high rate of rebleeding. For them hor-
monal therapy would be an answer.

So we looked at patients with small-bowel
angiodysplasias that we considered inoperable,
because they had too many lesions, too diffusely
scattered in the small intestine (15). We used
drugs similar to Bronner's: Enovid, a combination
of synthetic estrogen and progesterones, or the
naturally occurring estrogens in Premarin. There
were 64 patients that had diffuse angiodyspla-
sias; 30 patients received hormonal therapy and
34 did not. This is a cohort group; this is not ran-
domized; and the control group came from early
in our experience, before hormonal therapy really
became known. The two groups matched well in
sex and age; neither had a preponderance of as-
sociated illnesses like end-stage renal disease or
von Willebrand's; they had similar histories of
two years of bleeding; similar transfusion re-
quirements; and an average follow-up of over a
year in both groups.

Interestingly, both groups stopped bleeding.

Vol. 60 No. 3

OBSCURE GI BLEEDING— LEWIS

207

About half of the treated group and the untreated
group stopped bleeding in the follow-up period.
This data makes one rethink Bronner's data; he
had no control group and his stoppage rate was
57%. We concluded that hormonal therapy was
not effective in controlling bleeding from vascular
ectasias, especially patients without associated
renal failure or von Willebrand's or Osler-Weber-
Rendu disease.

Summary

I want to convey the notion that enteroscopy
has opened many doors, and continues to open up
more doors, in understanding and diagnosing dis-
eases of the small intestine. The true nature of
small-bowel angiodysplasia is still unanswered.
It seems unlikely that the lesions in the small
bowel are similar to the lesions that Scott Boley
talks about in the right colon. I doubt that the
intermittent obstruction to venous outflow, theo-
rized in the colon, is the pathophysiologic change
in the small intestine. Those studies, trying to
look for the changes that Boley described, need to
be done. We are trying to better characterize an-
giodysplasia of the small intestine, understand-
ing where they occur, with how many lesions, and
whether they are associated with any other ill-
nesses. We are looking at the association of small-
bowel vascular lesions with lesions in the stom-
ach and colon. Enteroscopy will in the future, we
hope, answer these questions. Enteroscopy, espe-
cially push enteroscopy, can help us with the
treatment of angiodysplasias.

We are now evaluating new instruments that
reach not just two feet beyond the ligament of
Treitz, but the entire jejunum, reaching 6 feet be-
yond the ligament of Treitz (16). Enteroscopy fa-
cilitates clinical research, can be used in patient
care, and guides treatment.

Questions and Answers

Sachar: We've heard about small-bowel enteros-
copy from the one person in the country who has
had more experience in this technique than any-
body else. Although lots of people have experience
with techniques, again it's to Blair's credit and
again in the Mount Sinai tradition that this ex-
perience hasn't been just a technique performed,
a technique that's been used, but has been stud-
ied, analyzed, and put to research and scholarly
uses while it is being used. Blair, I wanted to ask
you a sort of question for today and a question for
tomorrow.

Today, where does small-bowel enteroscopy
fit into that algorithmic sequence of studies? At

what point is it appropriate to do it? At what
point should other things be done first?

For tomorrow, when do you predict some
therapeutic potential will be built into the scope?
Lewis: Unfortunately, small-bowel enteroscopy is
not going to be ordered on admission along with
the complete blood count and the SMA 6. Enteros-
copy is for the patient who continues to bleed from
an unknown source, perhaps after two or three
colonoscopies have failed to yield a diagnosis. I
think that's appropriate. The yield increases with
increased numbers of examinations, to a point. I
think that small-bowel vascular ectasias are the
minority of the causes of bleeding. Cecal vascular
ectasias far outnumber small-bowel vascular
ectasias as a cause of bleeding. Examinations
may need to be repeated; that's number one.
Number two is that the patient's bleeding must
be really obscure — the patient must have an on-
going transfusion requirement. I think that en-
teroscopy is something that you consider when
you're thinking about doing an angiogram. I
think that our batting average is probably better
than angiography. So if you're thinking about do-
ing an angiogram, you should think about en-
teroscopy. That's today.

For tomorrow, the instrument is really a
state-of-the-art instrument. Although it would be
nice to have therapeutic modalities through this
instrument, when we tried to develop that type of
instrumentation, it became too thick and heavy.
It will not pass down through the entire small
intestine. So I don't think that's in the near fu-
ture. I think that the greatest hope is the new
push enteroscope I discussed; it can treat the en-
tire jejunum. That still has bugs in it, although
we've had some early experience here.
Sachar: Blair, you mentioned that in some pa-
tients multiple lesions have been found within
the small bowel. If you exclude Osier- Weber dis-
ease, do you find ectasias in other systems besides
the gut very often in these patients, and is that
ever a problem?

Lewis: These are not patients who have cerebral
ectasias or mucosal lesions or anything else that
points us in a direction. These are isolated le-
sions, and generally, the number of lesions most
people have is between two and five, and the le-
sions tend to be grouped together in one bit of
bowel. That just tends to be their course.
Levitt: Can you tell us something about the ra-
tionale for hormone therapy?
Lewis: The rationale is an old story going back to
Koch's 1952 report on a woman with Osier- We-
ber-Rendu disease who had terrible epistaxis that
varied with her menstrual cycle (17). People then

208

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

started to use hormonal therapy on epistaxis in
Osler-Weber-Rendu disease. In the ear-nose-
throat literature, many case reports suggested
success with this type of therapy. Finally in 1981
a randomized trial showed that the treatment
was ineffective (18). But the ball was then picked
up again, and there were several case reports on
the use of hormonal agents in people who were
bleeding from angiodysplasia of the bowel; then
came the works of Bronner and van Cutsem (13,
14).

How does it work? Simply, nobody knows.
Hormonal therapy is known to shorten bleeding
times in dialysis patients. In the nostrils of pa-
tients with Osler-Weber-Rendu disease, studies
show hormonal therapy thickens the mucus mem-
branes, putting a distance between the actual
blood vessel and the luminal orifice.
Sachar: I wonder if Dr. Levitt wants to know
what there is about end-stage renal disease that
makes these lesions different?
Levitt: We have a lot of this in our dialysis pop-
ulation. The blood pressure in those patients
bounces up and down. Is this a factor?
Lewis: Not that I'm aware of.
Sachar: Why does it happen in the stomach? Is
there a particular pathophysiology of the lesions
in end-stage renal disease?

Lewis: No one knows the pathophysiology of gas-
tric or small-bowel angiodysplasia. Boley's theory
that these lesions in the colon are related to in-
traluminal pressure has come under fire. Studies
looking at actual intraluminal pressures in the
cecum do not bear out his ideas.
Sachar: Are the lesions more frequent in renal
disease, or are they just more likely to bleed?
Lewis: Well, I think they are more common, be-
cause these lesions appear in patients who have
renal disease and not in normal individuals. The
incidence of cecal angiodysplasia runs between
2% and 3% in the population of normal persons
who have not had bleeding, based on endoscopic
studies and autopsy series. So the lesions occur
naturally in the colon, what Boley calls a common
degenerative change of aging, but they do not
necessarily bleed.

Sachar: Dr. Lewis left us in his talk with a note of
hope that the future of enteroscopy may answer
some of these questions. If it does, the answers
will come from him and from the Mount Sinai
team.

References

1. Boley S, Sammartano R, Adams A. On the nature and

etiology of vascular ectasias of the colon. Gastroenter-
ology 1977; 72:650-660.

2. Imperiale T, Ranschoff D. Aortic stenosis, idiopathic gas-

trointestinal bleeding and angiodyplasia: is there an
association? Gastroenterology 1988; 95:1670-1676.

3. Zuckerman G, Cornett G, Clouse R, Harter H. Upper gas-

trointestinal bleeding in patients with renal failure.
Ann Intern Med 1985; 102:588-592.

4. Fried A, Poulos A, Hatfield D. The effectiveness of the

incidental small bowel series. Radiology 1981; 140:45-
46.

5. Rex D, Lappas J, Maglinte D, Malczewski M, Kopecky K,

Cockerill E. Enteroclysis in the evaluation of suspected
small intestinal bleeding. Gastroenterology 1989; 97:
58-60.

6. Boley S, Sprayregen S, Sammartano R. The pathophysio-

logic basis for the angiographic signs of vascular ecta-
sias of the colon. Radiology 1977; 125:615-621.

7. Fiorito J, Brandt L, Kozicky O, Grosman I, Sprayragen S.

The diagnostic yield of superior mesenteric angiogra-
phy; Correlation with the pattern of gastrointestinal
bleeding. Am J Gastroenterol 1989; 84:878-881.

8. Retzlaff J, Hagedorn A, Bartholomen L. Abdominal explo-

ration for gastrointestinal bleeding of obscure origin.
JAMA 1961; 177:104-107.

9. Lewis B, Waye J. Small bowel enteroscopy for obscure GI

bleeding. Gastrointest Endosc 1991; 37:277 (abstr).

10. Lewis B, Kornbluth A, Waye J. Small bowel tumours,

yield of enteroscopy. Gut 1991; 32:763-765.

11. Biener A, Palestro C, Lewis B, Katz L. Intraoperative

scintigraphy for active small intestinal bleeding. Surg
Gynecol Obstet 1990; 171:388-392.

12. Lewis B, Wenger J, Waye J. Small bowel enteroscopy and

intraoperative enteroscopy for obscure gastrointestinal
bleeding. Am J Gastroenterol 1991; 86:171-174.

13. Bronner M, Pate M, Cunningham J, Marsh W. Estrogen-

progesterone therapy for bleeding gastrointestinal te-
langiectasias in chronic renal failure. Ann Intern Med
1986; 105:371-374.

14. Van Cutsem E, Rutgeerts P, Vantrappen G. Treatment of

bleeding gastrointestinal vascular malformations with
oestrogen-progesterone. Lancet 1990; 335:953-955.

15. Lewis B, Rivera-MacMurray S, Kornbluth A, Salomon P,

Waye J. Hormonal therapy for chronic GI bleeding from
diffuse small bowel AVMs, the results of a controlled
trial in 56 patients. Am J Gastroenterol 1990; 85:1266
(abstr).

16. Barkin J, Lewis B, Reiner D, Waye J, Goldberg R, Phillips

R. Diagnostic and therapeutic jejunoscopy with the SIF-
lOL enteroscope; longer is really better. Gastrointest
Endosc 1990; 36:214 (abstr).

17. Koch H, Escher G, Lewis J. Hormonal management of

hereditary hemorrhagic telangiectasia. JAMA 1952;
149:1376-1380.

18. Vase P. Estrogen treatment of hereditary hemorrhagic

telangiectasia. Acta Med Scand 1981; 209:393-396.

Grand
Rounds

The Pharmacology of
Antiinflammatory Agents:

A New Paradigm

Bruce N. Cronstein, M.D.

Inflammation is the final common response to a
variety of noxious stimuli, for example bacterial
invasion and trauma. Properly directed, inflam-
mation is the first step in wound healing. Al-
though beneficial with respect to elimination of
cellular debris or destruction of invading micro-
organisms, inflammation may, however, lead to
injury of host tissue. Indeed, joint destruction in
arthritis and much of the damage to the myocar-
dium incurred during myocardial infarction can
be ascribed to an appropriate inflammatory re-
sponse occurring at an inappropriate (or unfortu-
nate) site. The results of recent studies have led to
both an improved understanding of the factors
governing inflammation and, more importantly,
to new approaches to the control of inflammation
in such diseases as rheumatoid arthritis.

Inflammation is characterized by the accu-
mulation of leukocytes which, in the main, medi-
ate the inflammatory process. Recently we have
witnessed the biochemical dissection of inflam-
mation: a molecular understanding of those fac-
tors which attract leukocytes to sites of inflam-
mation and the biochemical events occurring in
leukocytes as they respond to inflammatory stim-
uli. Moreover, in the past six years we have be-
come aware of the central role of the endothelium
in directing leukocytes to specific sites of inflam-
mation.

Adapted from the author's Grand Rounds in Medicine presen-
tation at the Mount Sinai Medical Center on September 10,
1991. Final manuscript received June 1992. From the Depart-
ment of Medicine, Division of Rheumatology, New York Uni-
versity School of Medicine, New York. Address reprint re-
quests to the author, Associate Professor of Medicine, Division
of Rheumatology, New York University School of Medicine,
550 First Avenue, New York, NY 10016.

j The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

A large number of chemoattractants and in-
flammatory mediators have been the subject of
study. Among these mediators are the mast cell
product histamine, serum factors (complement
and the kinin and clotting systems), neurotrans-
mitters (substance P, ATP), lipids (platelet acti-
vating factor, lysolecithin, arachidonic acid), ei-
cosanoids (leukotrienes and prostaglandins),
growth factors (TGF-P) and, of great interest, the
family of proteins known as cytokines. Many of
the mediators listed above act as chemoattrac-
tants and direct stimuli for neutrophils; others
stimulate endothelium to produce chemoattrac-
tants. Recent studies have demonstrated that
many or all of the mediators listed above may be
targets for therapeutic intervention in the treat-
ment of inflammation.

As with other physiologic reactions, homeo-
stasis is maintained at the inflammatory site by
an appropriate balance of agonists and antago-
nists. At sites of inflammation cytokine antago-
nists, antiinflammatory prostaglandins (PGE^),
breakdown products of cellular constituents with
antiinflammatory activity (adenosine), and lipid
messengers with antiinflammatory activity (li-
poxins) are generated at sites of inflammation in
order to dampen the inflammatory response. Cur-
rent therapeutic approaches to the treatment of
inflammatory diseases such as rheumatoid ar-
thritis and inflammatory bowel disease make use
of these approaches. Thus, in recent studies we
have demonstrated that one potent antiinflam-
matory agent, methotrexate, is antiinflammatory
by virtue of its capacity to promote the release of
adenosine (1). Exogenous PGE^ is currently un-
der study for its use in the treatment of such in-
flammatory disorders as systemic lupus erythe-
matosus.

209

210

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

Neutrophils, the primary cells of acute in-
flammation, are short-lived cells that make their
way to sites of inflammation via the bloodstream.
These cells roll along the walls of the microvas-
culature until they encounter an appropriate at-
tractant or adhesive molecule. At that point they
adhere to the endothelium, migrate out between
endothelial cells into the extravascular space,
and migrate toward the inflammatory locus. For
many years it was thought that chemoattractants
generated in the extravascular space attracted
neutrophils to sites of inflammation. In this par-
adigm of inflammation the endothelium plays a
passive role in the direction of the inflammatory
process.

At sites of inflammation a variety of
chemoattractants are generated. Activation of
the complement system leads to generation of the
potent chemoattractant C5a. A variety of cells
metabolize arachidonic acid to bioactive deriva-
tives, such as LTB4 and HETEs, which may also
act as chemoattractants. Plasma membrane lip-
ids are converted to PAF, which also attracts neu-
trophils. More recently, cytokines (IL-8) and neu-
rotransmitters (Substance P) have also been
shown to be chemoattractants for neutrophils. In
general all of the substances listed above interact
with specific receptors on the neutrophil. Most of
these receptors have now been cloned and their
putative structure elucidated; they belong to a
family of G-protein-associated receptors (2). Reib-
man, in collaboration with my laboratory, has re-
cently demonstrated that TGFp is an extremely
potent chemoattractant for neutrophils and that
chemotaxis toward this molecule is not mediated
by a G-protein-related receptor (3).

The validity of the neutrophil-chemoat-
tractant model of inflammation depended on
the capacity of the neutrophil to increase its ad-
hesiveness for endothelium on contact with
chemoattractants. This suggested the hypothesis
that neutrophils must express or upregulate one
or more molecules that mediate adhesion of neu-
trophils to endothelium. The presence and impor-
tance of at least one of these molecules was first
demonstrated in the early 1980s as a result of a
combination of astute clinical investigation and
molecular biology. Several families were de-
scribed in which children inherited, in an autoso-
mal recessive fashion, a propensity to develop re-
current and severe bacterial infections. The
neutrophils and other leukocytes of these patients
were found to lack on their surface molecules
which, at the time, were known only as myeloid
differentiation antigens (CDlla,b,c/CD18, also
known as the (32 integrins). Subsequently, these

molecules were shown, on the neutrophil surface,
to be required for phagocytosis of complement-
opsonized particles and, more strikingly, for ad-
hesion to endothelium or other surfaces (4—7).
The (32 integrins are heterodimers with a com-
mon p-chain (CD18) which is not expressed in the
children suffering from leukocyte adhesion defi-
ciency (LAD). Subsequently other surface mole-
cules of the neutrophil have been shown to medi-
ate adhesion to endothelium as well (L-selectin or
LAM-1); moreover, L-selectin cooperates with
CDlla,b,c/CD18 during interaction of neutro-
phils with the surface of the endothelium (8-10).

Emigration of leukocytes into inflamed ex-
travascular loci occurs almost exclusively at the
postcapillary venule. The chemoattractant-leuko-
cyte model of inflammation does not explain the
exquisite localization of the inflammatory re-
sponse to the postcapillary venule. If the determi-
nant of neutrophil accumulation at sites of in-
flammation is the encounter between the
neutrophil and a chemoattractant, why should
one particular site in the microvasculature be the
only site at which leukocytes adhere and migrate
out into the extravascular space? In 1985 Bevil-
acqua and co-workers first demonstrated that cul-
tured endothelium exposed to IL-1 were more ad-
herent for resting neutrophils (11). This group
subsequently demonstrated that tumor necrosis
factor and endotoxin were equivalent with re-
spect to their capacity to induce adhesiveness of
endothelium and identified the endothelial adhe-
sive molecule (E-selectin or ELAM-1 [12-14]).
Studies performed in vitro demonstrated that, in
response to perfusion of an inflammatory agent,
E-selectin is expressed only on the endothelium of
the postcapillary venule (15). Thus the localiza-
tion to the postcapillary venule of the capacity to
respond to inflammatory stimuli explains why
the inflammatory response is channeled into the
extravascular space at a specific focus in the mi-
crovasculature.

The observation that endothelial cells could
express an adhesive molecule in response to in-
flammatory stimuli opened up an entirely new
area of research. Subsequently two other endo-
thelial molecules were shown to be expressed on
the surface of endothelium in response to inflam-
matory stimuli: molecules P-selectin (GMP-140)
and ICAM-1 (16-21). P-selectin, E-selectin, and
ICAM-1 are expressed or upregulated and then
down-regulated in a sequential fashion. P-selec-
tin, a molecule present in the granules of endo-
thelial cells, is expressed within minutes after ex-
posure to stimulus and disappears from the
surface by two hours after stimulation. E-selectin

Vol. 60 No. 3

ANTIINFLAMMATORY AGENTS— CRONSTEIN

211

is detectable on the surface of endothelium within
one hour of stimulation and disappears from the
endothelial surface by 12 hours. ICAM-1 is ex-
pressed constitutively and its surface expression
increases more slowly (18 hr) and is maintained
for longer periods (48 hr).

There are two other mechanisms by which
endothelial cells direct the inflammatory re-
sponse. Endothelial cells secrete the chemoattrac-
tant IL-8 into the extravascular space thus at-
tracting neutrophils out of the vasculature (22).
In addition the endothelium generates the chemo-
attractant PAF and expresses PAF on its surface
(23). Neutrophils adhere to and are stimulated by
the PAF on the surface of the endothelium, rein-
forcing the adhesion to P-selectin and E-selectin.
Thus, the endothelium has evolved several mech-
anisms designed to attract neutrophils to specific
inflamed sites.

The Mechanism of Action
Of Corticosteroids

Although corticosteroids have been used to
treat inflammatory disease for nearly four de-
cades the mechanism of action of these agents is
not completely understood. Various hypotheses
have been proposed to account for the antiinflam-
matory effects of steroids; these include "allo-
steric" effects on proteins (24), stabilization of ly-
sosomal and other cellular membranes (25, 26),
redirection of lymphocyte traffic (24, 27-33), di-
rect inhibition of various phospholipases (34), the
induction of such proteins as lipocortin (34—38),
and inhibition of the transcription of various cy-
tokines and metalloproteases (39-48). However,
none of these hypotheses completely account for
many of the pharmacologic effects of glucocorti-
coids: leukocytosis (49), inhibition of leukocyte re-
cruitment to inflamed areas (50, 51), retention of
lymphocytes in the lymphatic circulation with
shrinkage of peripheral lymph nodes (32, 33), and
the promotion of microbial infection (28, 33).

We therefore proposed the hypothesis that
corticosteroids are antiinflammatory by virtue of
their capacity to inhibit the expression by endo-
thelium of adhesive molecules upon stimulation
with endotoxin or cytokines. We found that dexa-
methasone at low concentrations (0.1-100 nM) in-
hibits the capacity of endotoxin-stimulated endo-
thelium to become more adhesive for neutrophils.
To confirm that dexamethasone was acting at its
receptor, we determined whether the steroid an-
tagonist RU-486 could reverse the effects of glu-
cocorticoids on stimulated endothelial adhesive-
ness (Fig. 1). Since RU-486, the potent steroid

antagonist, completely reversed the effects of
dexamethasone on stimulated endothelial adhe-
siveness, we concluded that dexamethasone was
acting at its receptor to inhibit endothelial adhe-
siveness. We next determined the effect of dexa-
methasone on expression of specific adhesogens
on the surface of the endothelium and observed
that dexamethasone inhibits the stimulated ex-
pression of E-selectin (ELAM-1) and upregulation
of ICAM-1 on endothelium. Again, RU-486 com-
pletely reversed the effects of dexamethasone on
E-selectin and ICAM-1 expression. Dexametha-
sone blocked the IL-1- and endotoxin-stimulated
accumulation of mRNA for E-selectin, suggesting
that glucocorticoids interfere, at the transcrip-
tional level, with expression of inflammatory ad-
hesogens on the surface of stimulated endothe-
lium. Surprisingly, dexamethasone did not
inhibit TNF-stimulated adhesiveness of endothe-
lium for neutrophils or the accumulation of
mRNA for E-selectin.

These studies indicate that steroids, in addi-
tion to their capacity to block the production of
inflammatory messengers such as cytokines,
blunt inflammation by inhibiting, at. the tran-
scriptional level, the upregulation of adhesogens
on the surface of endothelium.

The Mechanism of Action
Of Methotrexate

Methotrexate is a folate antagonist first in-
troduced for the treatment of malignancies. Very
soon after its introduction, methotrexate was
used in the therapy of an inflammatory disease,
rheumatoid arthritis (52); however, methotrexate
was not widely used for another thirty years, in
the early 1980s, when interest in this agent re-
vived. Low dose oral methotrexate has recently
been approved by the FDA for use in the treat-
ment of rheumatoid arthritis. Since approval for
use in the treatment of RA, methotrexate is now
one of the most commonly used second-line an-
tirheumatic agents (53).

Unlike most other antirheumatic agents,
methotrexate dramatically diminishes inflamma-
tion without diminishing the accumulation of in-
flammatory cells. Indeed, one recent study dem-
onstrates that there is a greater number of
neutrophils in the synovial fluid from patients
with rheumatoid arthritis being treated with
methotrexate (54). Moreover, unlike corticoste-
roids, methotrexate does not diminish the chronic
inflammatory infiltrate present in the synovium
of patients with rheumatoid arthritis. Despite the
presence of an inflammatory infiltrate similar in

cdNTROLt
%

•7 t

LPS + C^Rl ^ I

Fig. 1. Top panel Treatment of endothelial cells with endotoxin (LPS, 1 (xg/mL for 4 hr at 37°C) renders the
endothelium more adhesive for neutrophils (white arrow). Middle panel Treatment of endothelial cells
with endotoxin in the presence of sodium salicylate (SAL, 1 mM) or dexamethasone (DEX, 100 nM). Bottom
panel Treatment of endothelial cells with endotoxin in the presence of the inactive steroid analog tetrahy-
drocortisol (THC, 10 m-M) or Cortisol (CORT, 10 |xM).

Vol. 60 No. 3

ANTIINFLAMMATORY AGENTS— CRONSTEIN

213

composition and bulk to that present in the joints
of patients who are untreated, methotrexate
clearly ameliorates the inflammation of rheuma-
toid arthritis. Thus, methotrexate must diminish
the capacity of the cells of inflammation to induce
those changes in the joint which characterize
rheumatoid arthritis.

Despite widespread clinical use, the mecha-
nism of action of methotrexate has remained a
mystery. A number of clinical observations sug-
gest that methotrexate does not suppress inflam-
mation in rheumatoid arthritis by inhibition of
dihydrofolate reductase alone ("folate antago-
nism") with the consequent inhibition of purine
and pyrimidine biosynthesis. Unlike high-dose
methotrexate, the low doses of methotrexate used
to treat rheumatoid arthritis are rarely toxic to
the bone marrow. Indeed, bone marrow toxicity,
an expected side effect of high-dose methotrexate,
is an indication to stop or lower the dose of meth-
otrexate. Two recent blinded clinical studies dem-
onstrate that neither folate supplements nor cit-
rovorin rescue reverses the therapeutic eff'ects of
methotrexate in the therapy of rheumatic arthri-
tis (55, 56).

Recent advances in our understanding of the
metabolic fate of methotrexate have suggested an
explanation for the effectiveness of intermittent
administration of even low doses of methotrexate.
Methotrexate is rapidly taken up by cells and me-
tabolized to a series of polyglutamated deriva-
tives which persist intracellularly (57). Polyglu-
tamated methotrexate is an active inhibitor of
several enzymatically mediated steps involved in
synthesis and metabolism of purines and pyri-
midines (58-61). Of note, the enzyme most sensi-
tive to inhibition by polyglutamated methotrex-
ate is AICAR transformylase (Ki = 60nM) (59-
61) (Fig. 2).

We have found that methotrexate, presum-
ably via inhibition of AICAR transformylase and
accumulation of AICARibotide, induces release of
adenosine from fibroblasts and umbilical vein en-
dothelial cells. The doses of methotrexate re-
quired for promotion of adenosine release are well
within the range which can be achieved during
antirheumatic therapy of rheumatoid arthritis
(EC50 1 nM). Pretreatment of endothelial cells or
fibroblasts with methotrexate also inhibits, in
parallel, neutrophil adherence (stimulated and
unstimulated) to these connective tissue cells
(IC50 9 nM). Methotrexate inhibited neutrophil
adherence to connective tissue cells by increasing
adenosine release, since addition of adenosine de-
aminase to the supernatant medium completely
reversed the efi"ect of methotrexate pretreatment

on neutrophil adherence without, itself, afiecting
adherence (Fig. 2) (1). Thus, we showed that
methotrexate promotes release of adenosine from
connective tissue cells which inhibits both un-
stimulated and stimulated adherence of neutro-
phils to connective tissue cells. Our observations
suggest that the antiinflammatory actions of
methotrexate in vivo may also be mediated by
increased adenosine release at sites of inflamma-
tion.

At present the biochemical mechanism by
which methotrexate promotes adenosine release
remains a matter of speculation. Recent evidence
suggests that the accumulation of AICARibotide
intracellularly induces adenosine release. When
Gruber et al. (62) administered the cell-perme-
able, nonphosphorylated analog of AICARibotide,
AICARiboside, to dogs there was a marked rise in
adenosine release from ischemic cardiac tissue
that was postulated to be due to inhibition of
AMP deaminase. Subsequently, Barankiewicz
and colleagues have suggested that AICARibo-
side (and, therefore, intracellular AICARibotide)
inhibits adenosine utilization by cells (63). Our
observation that AICARiboside promotes adeno-
sine release and inhibits neutrophil-connective
tissue cell interactions is consistent with the hy-
pothesis that methotrexate inhibits the conver-
sion of AICARibotide to FAICARibotide by
AICARibotide transformylase and promotes the
intracellular accumulation of AICARibotide,
which, in turn, promotes adenosine release. We
have also examined the effect of AICARiboside on
both adenosine release from and neutrophil ad-
herence to either fibroblasts or endothelial cells.
We found that AICARiboside, like methotrexate,
also promoted adenosine release and inhibited
unstimulated and stimulated neutrophil adher-
ence to fibroblasts and endothelium. In a manner
identical to that of methotrexate, removal of ex-
tracellular adenosine by addition of exogenous
adenosine deaminase completely reversed the ef-
fect of AICARiboside on neutrophil adherence.
Thus, our results are consistent with the hypoth-
esis that methotrexate promotes adenosine re-
lease by interfering with intracellular metabo-
lism of AICARibotide (1).

The mechanism by which adenosine modu-
lates inflammatory cell function has been well es-
tablished; adenosine occupies specific receptors
on neutrophils, monocytes, and lymphocytes. In
general, as we first showed for the neutrophil, the
antiinflammatory and immunomodulatory effects
of adenosine are mediated by occupancy of A2 re-
ceptors (64—77). However, in vivo studies suggest,
based on the efficacy of receptor-specific analogs,

214

THE MOUNT SINAI JOURNAL OF MEDICINE

May 199

CONTROL METHOTREXATE AlCAR

Fig. 2. Top panel Treatment of endothelial cells with methotrexate (100 nM, 48 hr at 37°C) or AICARiboside (100 |xM, 2 h:
prevents adhesion of stimulated (FMLP, 100 nM) neutrophils to endothelium. Bottom panel Adenosine deaminase (0.12
lU/mL) reverses the effect of methotrexate pretreatment and AICARiboside treatment on adhesion of stimulated neutrophils 1
endothelium.

that adenosine and its analogs may engage
Ai receptors to suppress acute inflammation
(Fig. 3).

Thus, the studies described above demon-
strate a novel mechanism of action for the anti-
inflammatory drug methotrexate. Moreover, this
new understanding of the mechanism of action of
one of the most commonly used second-line agents
for the treatment of rheumatoid arthritis sug-
gests that new, more effective, and safer agents
can be developed for the treatment of rheumatoid
arthritis that take advantage of the local release

of antiinflammatory agents at sites of inflamma
tion.

Conclusion

The studies discussed here demonstrate the
one of the central events of inflammation is th
adhesion of neutrophils to endothelium, whicl
must precede the emigration of these inflamma
tory cells into the extravascular space. The mech
anism of action of two potent antiinflammator
agents, corticosteroids and methotrexate, are bes

Vol. 60 No. 3

ANTIINFLAMMATORY AGENTS— CRONSTEIN

215

rilli EFKIXTS OK ADKNOSINK KKCICITOK (KClll'ANCY
ON SIIMUIAIIJ) NiaJIROI'llll, HlN(TION

FUNCTION

Chcmdl.ixis

PhHgtJCylDsii

Superoxide iinion, ]\.0. . cil

Adhesion

Fig. 3.

explained as inhibition of neutrophil-endothelial
interactions by two different mechanisms. Corti-
costeroids blunt the capacity of the stimulated en-
dothelium to express adhesive molecules on its
surface, whereas methotrexate induces the endo-
thelium to release the potent antiinflammatory
autocoid adenosine.

Acknowledgments

These studies were performed during Dr. Cronstein's tenure
as the Irene Duggan Arthritis Investigator of the Arthritis
Foundation. The research was performed with the support of
grants from the American Heart Association (New York Af-
filiate), the Arthritis Foundation (New York Chapter) to Dr.
Cronstein, and from the National Institutes of Health (AR
11949, HL 1972) to Dr. Gerald Weissmann.

References

1. Cronstein BN, Eberle MA, Gruber HE, Levin RI, Meth-

otrexate inhibits neutrophil function by stimulating
adenosine release from connective tissue cells. Proc
Natl Acad Sci USA 1991; 88:2441-2445.

2. Lefkowitz RJ. Thrombin receptor: variations on a theme.

Nature 1991; 351:353-354.

3. Reibman J, Meixler S, Lee TC, et al. Transforming growth

factor beta 1, a potent chemoattractant for human neu-
trophils, bypasses classic signal-transduction path-
ways. Proc Natl Acad Sci USA 1991; 88:6805-6809.

4. Anderson DC, Schmalstieg FC, Arnaout MA, et al. Abnor-

malities of polymorphonuclear leukocyte function asso-
ciated with a heritable deficiency of high molecular
weight surface glycoprotein (GP 138); common relation-
ship to diminished cell adherence. J Clin Invest 1984;
74:536-551.

5. Springer TA, Thompson WS, Miller LJ, Schmalstieg FC,

Anderson DC. Inherited deficiency of the Mac-1, LFA-1
and gp 150/95 glycoprotein family and its molecular ba-
sis. J Exp Med 1984; 160:1901-1918.

6. Sanchez-Madrid F, Nagy JA, Robbins E, Simon P,

Springer TA. A human leukocyte differentiation anti-
gen family with distinct alpha subunits and a common
beta subunit: the lymphocyte function associated anti-
gen (LFA-1), the iC3b complement receptor (OKMl/
Mac-1), and the pl50/95 molecule. J Exp Med 1983;
158:1785-1803.

7. Arnaout MA, Spits H, Terhorst C, Pitt J, Todd RF III.

Deficiency of a leukocyte surface glycoprotein (LFA-1)
in two patients with Mol deficiency: effects of cell acti-
vation on Mol/LFA-1 surface expression in normal and
deficient leukocytes. J Clin Invest 1984; 74:1291-1300.

8. Von Andrian UH, Chambers JD, McEvoy LM, Bargatze

RF, Arfors KE, Butcher EC. Two-step model of leuko-

cyte-endothelial cell interaction in infiammation: dis-
tinct roles for LECAM-1 and the leukocyte beta 2 inte-
grins in vivo. Proc Natl Acad Sci USA 1991; 88:7538-
7542.

9. Smith CW, Kishimoto TK, Abbass O, et al. Chemolactic
factors regulate lectin adhesion molecule 1 (LEC AM-
D-dependent neutrophil adhesion to cytokine-stimu-
lated endothelial cells in vitro. J Clin Invest 1991; 87:
609-618.

10. Smith CW, Marlin SD, Rothlein R, Toman C, Anderson

DC. Cooperative interactions of LFA-1 and Mac-1 with
intercellular adhesion molecule-1 in facilitating adher-
ence and transendothelial migration of human neutro-
phils in vitro. J Clin Invest 1991; 83:2008-2017.

11. Bevilacqua MP, Pober JS, Wheeler ME, Cotran RS, Gim-

brone MA. Interleukin 1 acts on cultured human endo-
thelium to increase the adhesion of polymorphonuclear
leukocytes, monocytes, and related leukocyte cell lines.
J Clin Invest 1985; 76:2003-2011.

12. Pober JS, Bevilacqua MP, Mendrick DL, Lapierre LA, Pi-

ers W, Gimbrone MA Jr. Two distinct monokines, in-
terleukin 1 and tumor necrosis factor, each indepen-
dently induce biosynthesis and transient expression of
the same antigen on the surface of cultured human vas-
cular endothelial cells. J Immunol 1986; 136:1680-
1687.

13. Pober JS, Lapierre LA, Stolpen AH, et al. Activation of

cultured human endothelial cells by recombinant lym-
photoxin: comparison with tumor necrosis factor and
interleukin 1. J Immunol 1987; 138:3319-3324.

14. Bevilacqua MP, Pober JS, Mendrick DL, Cotran RS, Gim-

brone MA. Identification of an inducible endothelial-
leukocyte adhesion molecule. Proc Natl Acad Sci USA
1987; 84:9238-9242.

15. Messadi DV, Pober JS, Fiers W, Gimbrone MA Jr., Mur-

phy GF. Induction of an activation antigen on postcap-
illary venular endothelium in human skin organ cul-
ture. J Immunol 1987; 139:1557-1562.

16. Rothlein R, Dustin ML, Marlin SD, Springer TA. A hu-

man intercellular adhesion molecule (ICAM-1) distinct
from LFA-1. J Immunol 1986; 137:1270-1274.

17. Dustin ML, Rothlein R, Bhan AK, Dinarello CA, Springer

TA. Induction by IL-1 and interferon-gamma: tissue
distribution, biochemistry, and function of a natural ad-
herence molecule (ICAM-1). J Immunol 1986; 137:245-
254.

18. Smith CW, Rothlein R, Hughes BJ, et al. Recognition of

an endothelial determinant for CD18-dependent hu-
man neutrophil adherence and transendothelial migra-
tion. J Clin Invest 1988; 82:1746-1756.

19. Marlin SD, Springer TA. Purified intercellular adhesion

molecule-1 (ICAM-1) is a ligand for lymphocyte func-
tion-associated antigen 1 (LFA-1). Cell 1987; 51:813-
819.

20. Dustin ML, Springer TA. Lymphocyte function-associated

antigen-1 (LFA-1) interaction with intercellular adhe-
sion molecule-1 (ICAM-1) is one of at least three mech-
anisms for lymphocyte adhesion to cultured endothelial
cells. J Cell Biol 1988; 107:321-331.

21. Geng JG, Bevilacqua MP, Moore KL, et al. Rapid neutro-

phil adhesion to activated endothelium mediated by
GMP-140. Nature 1990; 343:757-760.

22. Huber AR, Kunkel SL, Todd RF III, Weiss SJ. Regulation

of transendothelial neutrophil migration by endoge-
nous Interleukin-8. Science 1991; 254:99-102.

23. Zimmerman GA, Mclntyre TM, Mehra M, Prescott SM.

Endothelial cell-associated platelet-activating factor: a

216

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

novel mechanism for signaling intercellular adhesion. J
Cell Biol 1990; 110:529-540.

24. Samuels H, Tomkins GM. Relation of steroid structure to

enzyme induction in hepatoma tissue culture cells. J
Mol Biol 1970; 52:57-74.

25. Weissmann G, Thomas L. Studies on lysosomes: II. the

effect of cortisone on the release of acid hydrolases from
a large granule fraction of rabbit liver induced by an
excess of vitamin A. J Clin Invest 1963; 42:661-669.

26. Bangham AD, Standish MM, Weissmann G. The action of

steroids and streptolysin S on the permeability of phos-
pholipid structures to cations. J Mol Biol 1965; 13:253-
259.

27. Fauci AS, Dale DC. The effect of hydrocortisone on the

kinetics of normal human lymphocytes. Blood 1975;
46:235-243.

28. Fauci AS, Dale DC. The effect of in vivo hydrocortisone on

subpopulations of human lymphocytes. J Clin Invest
1974; 53:240-246.

29. Fauci AS, Dale DC, Balow JE. Glucocorticosteroid ther-

apy: mechanisms of action and clinical considerations.
Ann Intern Med 1976; 84:304-315.

30. Fauci AS. Human bone marrow lymphocyte: I. Distribu-

tion of lymphocyte subpopulations in the bone marrow
of normal individuals. J Clin Invest 1975; 56:98-103.

31. Dale DC, Fauci AS, Dupont G IV, Wolff SM. Comparison

of agents producing a neutrophilic leukocytosis in man:
hydrocortisone, prednisone, endotoxin and etiochol-
anolone. J Clin Invest 1975; 56:808-813.

32. Fauci AS, Dale DC. The effect of hydrocortisone on the

kinetics of normal human lymphocytes. Blood 1975; 46:
235-243.

33. Parillo JE, Fauci AS. Mechanisms of glucocorticoid action

on immune processes. Ann Rev Pharmacol Toxicol
1979; 19:179-201.

34. Blackwell GJ, Flower RJ, Nijkamp FP, Vane JR. Phos-

pholipase A2 activity of guinea-pig isolated perfused
lungs: stimulation and inhibition by anti-inflammatory
steroids. Br J Pharmacol 1978; 62:79-89.

35. Rosa MD, Flower RJ, Hirata F, Parente L, Russo-Marie F.

Anti-phospholipase proteins. Prostaglandins 1984; 28:
441^42.

36. Larsen C, Claus Z, Mukaida N, et al. Cytokines and lipo-

cortins in inflammation and differentiation. New York:
- Wiley-Liss, 1990.

37. DiRosa M, Flower RJ, Hirata F, Parenti L, Russo-Marie F.

Anti-phospholipase proteins. Prostaglandins 1984; 28:
441-442.

38. Flower RJ. Background and discovery of lipocortins.

Agents and Actions 1985; 17:255-262.

39. Bochner BS, Rutledge BK, Schleimer RP. Interleukin 1

production by human lung tissue: II. Inhibition by anti-
inflammatory steroids. J Immunol 1987; 139:2303-
2307.

40. Djaldetti R, Fishman P, Shtatlender V, Sredni B, Djal-

detti M. Effect of dexamethasone on IL-1 and IL-3-LA
release by unstimulated human mononuclear cells.
Biomed Pharmacother 1990; 44:515-518.

41. Kimbauer R, Kock A, Neuner P, et al. Regulation of epi-

dermal cell interleukin-6 production by UV light and
corticosteroids. J Invest Dermatol 1991; 96:484-^89.

42. Chensue SW, Terebuh PD, Remick DG, Scales WE,

Kunkel SL. In vivo biologic and immunohistochemical
analysis of interleukin- 1 alpha, beta and tumor necro-
sis factor during experimental endotoxemia: kinetics,
Kupffer cell expression, and glucocorticoid effects. Am J
Pathol 1991; 138:395-402.

43. Zuckerman SH, Shellhaas J, Butler LD. Differential reg-

ulation of lipopolysaccharide-induced interleukin 1 and
tumor necrosis factor synthesis: effects of endogenous
and exogenous glucocorticoids and the role of the pitu-
itary-adrenal axis. Eur J Immunol 1989; 19:301-305.

44. Knudsen PJ, Dinarello CA, Strom TB. Glucocorticoids in-

hibit transcriptional and post-transcriptional expres-
sion of interleukin 1 in U937 cells. J Immunol 1987;
139:4129-4134.

45. Dinarello CA, Mier JW. Lymphokines. N Engl J Med

1987; 317:940-945.

46. Beutler B, Cerami A. The biology of cachectin/TNF — a

primary mediator of the host response. Ann Rev Immu-
nol 1989; 7:625-655.

47. Shapiro SD, Campbell EJ, Kobayashi DK, Welgus HG.

Dexamethasone selectively modulates basal and lipo-
polysaccharide-induced metalloproteinase and tissue
inhibitor of metalloproteinase production by human al-
veolar macrophages. J Immunol 1991; 146:2724-2729,

48. Clark SD, Kobayashi DK, Welgus HG. Regulation of the

expression of tissue inhibitor of metalloproteinases and
coUagenase by retinoids and glucocorticoids in human
fibroblasts. J Clin Invest 1987; 80:1280-1288.

49. Athens JW, Haab OP, Raab SO, Mauer AM, Ashen-

brucker H, Cartwright GE, Wintrobe MM. Leukoki-
netic studies: IV. The total blood, circulating and mar-
ginal granulocyte pools and the granulocyte turnover
rate in normal subjects. J Clin Invest 1961; 40:989-995.

50. Selye H. On the mechanism through which hydrocorti-

sone affects the resistance of tissue to injury. JAMA
1953; 152:1207-1213.

51. Zweiman B, Slott RI, Atkins PC. Histologic studies of hu-

man skin test responses to ragwood and compound 48/
80. III. Effect of alternate day steroid therapy. J Allergy
Clin Immunol 1976; 58:657-663.

52. Gubner R, August S, Ginsberg V. Therapeutic suppres-

sion of tissue reactivity: II. Effect of aminopterin in
rheumatoid arthritis and psoriasis. Am J Med Sci 1951;
221:176-182.

53. Furst DE, Kremer JM. Methotrexate in rheumatoid ar-

thritis. Arthritis Rheum 1988; 31:305-314.

54. Bahremand M, Schumacher HR Jr. Effect of medication

on synovial leukocyte differentials in patients with
rheumatoid arthritis. Arthritis Rheum 1991; 34:1173-
1176.

55. Hanrahan PS, Russell AS. Concurrent use of folinic acid

and methotrexate in rheumatoid arthritis. J Rheumatol
1988; 15:1078-1080.

56. Morgan SL, Baggott JE, Vaughn WH, Young PK, Austin

JV, Krumdieck CL, Alarcon GS. The effect of folic acid
supplementation on the toxicity of low-dose methotrex-
ate in patients with rheumatoid arthritis. Arthritis
Rheum 1990; 33:9-18.

57. Chabner BA, Allegra CJ, Curt GA, et al. Polyglutamation

of methotrexate: Is methotrexate a prodrug? J Clin In-
vest 1985; 76:907-912.

58. Zimmerman CL, Franz TJ, Slattery JT. Pharmacokinetics

of the poly-gamma-glutamyl metabolites of methotrex-
ate in skin and other tissues of rats and hairless mice.
J Pharmacol Exp Ther 1984; 231:242-247.

59. Allegra CJ, Drake JC, Jolivet J, Chabner BA. Inhibition

of phosphoribosylaminoimidazolecarboxamide trans-
formylase by methotrexate and dihydrofolic acid poly-
glutamates. Proc Natl Acad Sci USA 1985; 82:4881-
4885.

60. Allegra CJ, Hoang K, Yah GC, Drake JC, Baram J. Evi-

dence for direct inhibition of de novo purine synthesis

Vol. 60 No. 3

ANTIINFLAMMATORY AGENTS— CRONSTEIN

217

in human MCF-7 breast cells as a principal mode of
metabolic inhibition by methotrexate. J Biol Chem
1987; 262:13520-13526.

61. Baggott JE, Vaughn WH, Hudson BB. Inhibition of 5-ami-

noimidazole-4-carboxamide ribotide transformylase,
adenosine deaminase and 5'-adenylate deaminase by
polyglutamates of methotrexate and oxidized folates
and by 5-aminoimidazole-4-carboxamide riboside and
ribotide. Biochem J 1986; 236:193-200.

62. Gruber HE, Hoffer ME, McAllister DR, Laikind PK, Lane

TA, Schmid-Schoenbein GW, Engler RL. Increased
adenosine concentration in blood from ischemic myo-
cardium by AICA riboside: effects on flow, granulocytes
and injury. Circulation 1989; 80:1400-1411.

63. Barankiewicz J, Ronlov G, Jimenez R, Gruber HE. Selec-

tive adenosine release from human B but not T lym-
phoid cell line. J Biol Chem 1990; 265:15738-15743.

64. Cronstein BN, Rosenstein ED, Kramer SB, Weissmann G,

Hirschhorn R. Adenosine; a physiologic modulator of
superoxide anion generation by human neutrophils:
adenosine acts via an A2 receptor on human neutro-
phils. J Immunol 1985; 135:1366-1371.

65. Schmeichel CJ, Thomas LL. Methylxanthine bronchodi-

lators potentiate multiple human neutrophil functions.
J Immunol 1987; 138:1896-1903.

66. Schrier DJ, Imre KM. The effects of adenosine agonists on

human neutrophil function. J Immunol 1986; 137:
3284-3289.

67. Cronstein BN, Kramer SB, Rosenstein ED, Korchak HM,

Weissman G, Hirschhorn R. Occupancy of adenosine
receptors raises cyclic AMP alone and in synergy with
occupancy of chemoattractant receptors and inhibits
membrane depolarization. Biochem J 1988; 252:709-
715.

68. De la Harpe J, Nathan CF. Adenosine regulates the re-

spiratory burst of cytokine-triggered human neutro-
phils adherent to biologic surfaces. J Immunol 1989;
143:596-602.

69. Nielson CP, Vestal RE. Effects of adenosine on polymor-

phonuclear leucocyte function, cyclic 3': 5'-adeno8ine
monophosphate, and intracellular calcium. Br J Phar-
macol 1989; 97:882-888.

70. Roberts PA, Newby AC, Hallett MB, Campbell AK. Inhi-

bition by adenosine of reactive oxygen metabolite pro-
duction by human polymorphonuclear luecocytes. Bio-
chem J 1985; 227:669-674.

71. Eppell BA, Newell AM, Brown EJ. Adenosine receptors

are expressed during differentiation of monocytes to
macrophages in vitro. J Immunol 1989; 143:4141—4145.

72. Lappin D, Whaley K. Adenosine A2 receptors on human

monocytes modulate C2 production. Clin Exp Immunol
1984; 57:454-460.

73. Riches DWH, Watkins JL, Henson PM, Stanworth DR.

Regulation of macrophage lysosomal secretion by aden-
osine, adenosine phosphate esters, and related struc-
tural analogues of adenosine. J Leuk Biol 1985; 37:545-
557.

74. Marone G, Vigorita S, Triggiani M, Condorelli M. Aden-

osine receptors on human lymphocytes. Adv Exp Med
Biol 1986; 195:7-14.

75. Bonnafous J-C, Dornand J, Favero J, Mani J-C. Lympho-

cyte membrane adenosine receptors coupled to adenyl-
ate cyclase; properties and occurrence in various lym-
phocyte subclasses. J Receptor Res 1981; 2:347-366.

76. Mandler R, Birch RE, Polmar SH, Kammer GM, Rudolph

SA. Abnormal adenosine-induced immunosuppression
and cAMP metabolism in T lymphocytes of patients
with systemic lupus erythematosus. Proc Natl Acad Sci
USA 1982; 79:7542-7546.

77. Salmon JE, Cronstein BN. Fcgamma receptor-mediated

functions in neutrophils are modulated by adenosine
receptor occupancy: Al receptors are stimulatory and
A2 receptors are inhibitory. J Immunol 1990; 145:
2235-2240.

Grand
Rounds

Asthma:

An Inflammatory Disease

Kirk Sperber, M.D.

Asthma has been thought of as a reversible ob-
structive lung disease in which inflammation
played little or no role. Over the past 15 years,
however, it has become evident that asthma is a
chronic inflammatory disease (1), similar to the
rheumatologic disorders, which contributes both
directly and indirectly to the obstructive process.
The mechanisms leading to this inflammation are
multiple and complex. The presence of inflamma-
tory cells and their mediators in the airways pro-
duce alterations in bronchial epithelial cells that
lead to their destruction and to increased smooth-
muscle hyperreactivity characteristic of asthma
(2, 3). Evidence for the prominent role of inflam-
mation in asthma comes from morphologic stud-
ies demonstrating the presence of inflammatory
cells, including neutrophils, mononuclear cells,
and especially eosinophils, in the airway associ-
ated with mucosal edema (4). Other investigators
have demonstrated increased numbers of eosino-
phils and mast cells in bronchoalveolar fluid
(BALF) (5) from asthmatics during mild flareups
of disease and increases in these cells in the spu-
tum of asthmatics during mild exacerbations (6).
Even in patients who have the mildest asthma,
the mucosa has been demonstrated to be abnor-
mal, showing mast-cell degranulation with wide-
spread eosinophilic infiltration coupled with in-
creases in mucosal neutrophils and mononuclear
cells (7, 8).

If asthma is an inflammatory disease and if

Adapted from the author's Grand Rounds presentation on Oc-
tober 1, 1991. Final manuscript revision received May 1992.
From the Division of Clinical Immunology, Department of
Medicine, The Mount Sinai Medical Center, New York City.
Address all correspondence and reprint requests to the author
at Box 1089, Division of Clinical Immunology, Department of
Medicine, The Mount Sinai Medical Center, One Gustave L.
Levy Place, New York, NY, 10029.

218

proinflammatory cellular infiltration is a disease
characteristic, how is the presence of these cells
responsible for the onset of disease? Proinflam-
matory cells like eosinophils and mast cells, when
activated, release both stored and newly synthe-
sized mediators, including cytokines, which have
a wide variety of biological activities (9). These
mediators are responsible for the clinical features
of asthma, including bronchoconstriction, micro-
vascular leakage, edema, exudation, and mucus
hypersecretion (10-12).

Atopy and Asthma

Although asthma was once considered a dis-
ease occasionally associated with allergic factors,
several recent large studies have clearly estab-
lished a connection between atopy and asthma in
a large percentage of asthmatics. These studies
have come primarily from rural and suburban
populations and have associated either elevated
IgE levels or positive radioallergosorbent tests
(RASTs) to a limited number of aeroallergens
with emergency visits for asthma treatment (IS-
IS). Burrows et al. (16) in a large Arizona study
found higher IgE levels in 90% of patients below
the age of 40 reporting to local emergency rooms
for asthma treatment than in age-matched non-
asthmatic controls. Similarly, Pollart et al. (17)
and Reid et al. (18) in two studies, one in rural
Virginia and one in northern California, associ-
ated increased emergency visits for asthma in
adult patients below the age of 50 with positive
RASTs for dust mites and rye grass respectively.

Although the role of atopy and asthma has
been investigated in rural and suburban popula-
tions, there are relatively few data regarding the
role of atopy in inner-city urban asthmatic popu-
lations. At The Mount Sinai Hospital in New
York City, N.Y., we have investigated the role of

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

ASTHMA AS INFLAMMATORY— SPERBER

219

atopy and asthma in asthmatics coming to the
emergency room (19). The mean IgE level of asth-
matic patients (n = 73) reporting to the Mount
Sinai Emergency Room was 263.8 lU/mL, com-
pared to 63.8 lU/mL in the nonasthmatic control
group (p < 0.032) (n = 30). Sixty-eight percent of
the asthmatics had an IgE level in the atopic
range (>100 lU), which is consistent with other
data nationwide.

In contrast to the studies of rural and subur-
ban populations, however, the urban asthmatic
group we investigated had IgE antibodies di-
rected against an indoor panel of allergens as de-
termined by RAST. Our population was allergic
to a predominantly indoor panel of allergens,
which provide constant exposure to allergens
leading to chronic airway inflammation, includ-
ing cockroach, cat, dog, and dust mite. This con-
trasts with studies (13-18) which had demon-
strated either IgE antibodies or positive allergy
skin tests to grass pollen, ragweed pollen, tree
pollen, and mold pollen.

Mast Cells and Asthma

Mast cell populations are not homogeneous,
and two major types of mast cells have been de-
scribed, the mucosal T-type mast cell (MCx) and
the CT-type connective tissue (MC^t) niast cell
(20). These cells differ from each other in a vari-
ety of ways, including enzyme content and loca-
tion (Table 1). MC^ cells contain tryptase (hence
the name T-type); MC^t contain both tryptase
and chymase (hence the name CT) (21). MCj are
present in the alveolar wall, bronchi, bronchioles,
and nasal and bowel mucosa, whereas MC^t are
present in skin, intestinal submucosa, and nasal
mucosa (22). Although one type of mast cell may
predominate at a particular anatomic site, the
other type of mast cell may also be present (22).
MCx MCcT also differ in the lattice structure
of their cytoplasmically stored preformed media-
tors; the MCx type mast-cell mediators have a
scroll-like appearance, whereas MCcx preformed
mediators have a lattice structure with a grading
crystal formation (23). MCx cells are T-cell depen-
dent and their numbers are dramatically higher
in parasitic infections and in asthma than in the
MCcx (^24). In contrast, in T-cell-deficient states,
either acquired immunodeficient states or pri-
mary T-cell disorders, MCx absent (25). An-
other differentiating point is histamine content.
Connective mast cells contain more histamine
than the mucosal mast cells (26) and release his-
tamine more readily, in response not only to IgE
crosslinking but also to other stimuli, including

TABLE 1

Comparison of T -Type (MC-,-) and CT-Type Connective
Tissue (MCcT> l^ast Cell

Characteristic

MCt

MCcT

Tissue distribution

Skin

+ +

Intestinal submucosa

+ +

Intestinal mucosa

+ +

Alveolar wall

Bronchi/bronchioles

+ +

Nasal mucosa

+ +

Conjunctiva

+ +

Enzyme content

Tryptase

Chymase

T-cell dependency

Yes

Granule morphology

Grating, lattice

Scroll

compound 48/80, codeine, and radiocontrast dye
(27).

Looking at the development of the two mast
cell types, it is uncertain whether they are de-
rived from a single uncommitted stem cell which
can then become either a MCx MC^x or wheth-
er they are derived as separate lineages. Mast
cells, like basophils, are derived from bone-mar-
row stem cells (28), but in contrast to other bone-
marrow-derived cells, they undergo only partial
differentiation in the marrow and complete the
differentiation process in peripheral-mucosal and
connective-tissue microenvironments (29). MCx
and MCcx arise from hematopoietic progenitor
stem cells (28). The progenitor cells with commit-
ment to a mast-cell lineage are released into the
circulation. At the time of commitment and soon
thereafter, they enter their tissue site of resi-
dence and develop along MCcx or MCx pathways
to the final mature forms. Whether the predomi-
nance of a particular type of mast cell in different
tissue is due to selective recruitment of precom-
mitted mast cells to a particular site is unclear
(29). T-lymphocyte-associated activity is impor-
tant for MCx cells, whereas fibroblast-derived fac-
tors appear to be important in both mast cell
types (30).

Allergen-induced mast cell activation
through crosslinking of IgE antibodies bound to a
high-affinity Fc receptor (FcEl) is associated with
both an early phase reaction characterized by the
release of preformed mediators, histamine, and
the proteolytic enzymes tryptase and chymase,
and a late phase reaction (LPR) characterized by
generation of proinflammatory mediators (31).
The late phase reaction is associated with the re-
cruitment of eosinophils and neutrophils into the
airways and is more important than the early

220

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

TABLE 2

Pharmacologic Effects of Major Cyclooxygenase and
Lipooxygenase Products in Asthma

Mediator

Pharmacologic action

Cyclooxygenase
PGD2

Lipooxygenase
LTB4

LTC4
LTD4
LTE.

Bronchoconstrictor
Peripheral vasodilator
Neutrophil chemotaxis
Tussive agent
Bronchoconstrictor
Peripheral vasodilator
Coronary vasoconstrictor

Neutrophil and eosinophil

Chemotaxis and adherence

Vasopermeability

Bronchoconstrictor

Vasopermeability

Bronchoconstrictor

Vasopermeability

phase reaction in propagating airways inflamma-
tion (32).

The presence of such a wide variety of medi-
ators has therapeutic imphcations. For example,
although histamine levels are elevated in serum
and bronchoalveolar lavage fluid (BALF) of asth-
matics, antihistamine therapy in general has
been unsuccessful in the treatment of asthma
(33). The failure of antihistamines as antiasth-
matic agents is not surprising given the large
number of mediators involved in this process. The
presence of so many mediators makes it unlikely
that inhibition of any single one will be sufficient
to treat all of the symptoms of asthma.

Mast cell death does not occur from IgE
crosslinking activation, and in 24 hours pre-
formed mediators are resynthesized and are
present in intracytoplasmic stores ready to be re-
released after subsequent restimulation (34). Im-
mediate mast cell degranulation can be demon-
strated clinically by allergy skin testing, in which
exposure of allergen to MC^x cells sensitized with
allergen-specific IgE results in histamine release.
Late phase reactions can also occur with allergy
skin testing but generally occur only about 25% of
the time for a given dose of allergen. With larger
doses of allergen administrated in the early re-
sponse, LPRs will occur 100% of the time (35).
Biopsies obtained from the early response and the
late reaction are dramatically different; the early
response yield is relatively acellular (36),
whereas the late response yield is associated with
the accumulation of large numbers of eosinophils
and neutrophils (37). The same inflammatory

events with cellular accumulation of eosinophils
and neutrophils that occur in the skin after skin
testing also occur in the airway after allergen ex-
posure.

Synthesized Lipid Mediators

The cyclooxygenase products play an impor-
tant role in the pathogenesis of asthma. These
mediators are capable of causing bronchoconstric-
tion and bronchohyperresponsiveness (Table 2),
increased mucus secretion, and microvascular
leak, which can cause bronchial obstruction (38).
They can cause cough interacting directly with
tussive receptors in the airways. The major cyclo-
oxygenase products implicated in asthma are
PGD2 and PGFaa (39), which are 30 times more
potent than histamine as bronchoconstrictors
(40). Elevated levels of these mediators have been
demonstrated in serum of both chronic asthmat-
ics and in BALF from asthmatics after allergen
challenge (41). Treatment with cyclooxygenase
inhibitors, including aspirin, have only limited
efficacy in treating chronic asthma and may even
cause exacerbation by as yet undetermined mech-
anisms (42). The exact role of cyclooxygenase
products in asthma is still somewhat uncertain
because of the lack of effective inhibitors at more
specific sites in the biosynthetic pathway. The de-
velopment of more specific inhibitors will help
elucidate the role of these products in the patho-
genesis of asthma.

Leukotriene products can similarly cause air-
way smooth-muscle contraction, airway hyperre-
sponsiveness, mucus release, and microvascular
leak (Table 2) and are among the most potent
chemotoxins known for neutrophils and eosino-
phils (43). LTC4, LTD4, LTE4, and LTB4 are the
principal leukotrienes associated with asthma
(44). As bronchoconstrictors, the leukotrienes are
one thousand times more potent than histamine
and can potentiate the bronchoconstrictive effects
of histamine (45). The precise role of leukotriene
products in asthma, like the role of cyclooxygen-
ase products, is unclear because of the lack of spe-
cific inhibitors. However, clinical trials with more
promising agents have given encouraging prelim-
inary results (46, 47) which may lead to more
widespread use of these agents.

Platelet activating factor (PAF) was so
named because it was first derived from IgE-sen-
sitized rabbit lung tissue and was demonstrated
to cause aggregation of platelets (48). PAF is gen-
erated by phospholipase A2, the same enzyme
that synthesizes the cyclooxygenase and lipooxy-
genase products. Instead of producing precursors
for arachidonic acid and leukotriene products,

Vol. 60 No. 3

ASTHMA AS INFLAMMATORY— SPERBER

221

lyso-PAF is synthesized, and can then be further
metabolized into PAF (49). PAF production is not
Hmited to mast cells; macrophages can also pro-
duce significant amounts of this mediator (50).
PAF in the airways is a potent bronchoconstric-
tor, both in normal persons and asthmatics, and if
injected in the skin can cause a wheel-and-flare
reaction (51). In experimental animal model sys-
tems, PAF also causes microvascular leakage,
mucus secretion, and edema, which may contrib-
ute to airway obstruction (52).

PAF interacts with specific receptors on a va-
riety of cells, including neutrophils and eosino-
phils (53), and — like the leukotrienes — can cause
activation and chemotaxis of eosinophils in vivo
and in vitro (54). Activated eosinophils produce
several proteins, including major basic protein
(MBP), eosinophilic cationic protein (ECP), and
eosinophil-derived neurotoxin (EDN), which di-
rectly damage the respiratory epithelium (55).
PAF, produced either by a macrophage or a mast
cell, activates eosinophils to release MBP, ECP,
and EDN. The interaction of these proteins with
the epithelial cells causes disruption and desqua-
mation and exposes subbronchial unmyelinated
fibers (56). Stimulation of the exposed axons by
reflex action causes increased bronchial hyperre-
sponsiveness (57). The loss of epithelial cells also
results in the loss of an epithelial-cell-derived
smooth-muscle relaxing factor which can further
contribute to bronchial hyperreactivity (58).

Mast cell activation and elaboration of medi-
ators could explain many of the features of
asthma. Activation of mast cells through
crosslinking of IgE results in the release of the
preformed and newly synthesized mediators,
which cause vasodilatation and exudation, ac-
companied by the infiltration of eosinophils and
neutrophils into the site of allergen exposure. At
the site of inflammation the neutrophils and
eosinophils are activated and release MBP, EDN,
and ECP, which interact with the epithelial cells,
causing desquamation and exposing the sensory
nerve fibers, increasing the muscular tone and
the reactivity of the bronchial smooth muscle
(Fig.).

Although this model system is appealing, it
does not account for a substantial part of the
chronic inflammation present in asthma.

T Cells and Asthma

It has become apparent over the past two
years that T cells, and the cytokines produced by
them, are important cells in propagating the
chronic inflammatory process in asthma (59, 60).

Vasodilation, LeukocvTo Recaplor Mediator

exudation marginaiion Chemotaxi* enhancement release

Bronctioconstriction Nerve Mucus Epithelial

stimulation secretion desquamation

Fig. Schematic view of role of mast and mast cell mediators
in pathogenesis of allergic airway inflammation.

Evidence for the role of T cells in asthma comes
from several studies in which activated periph-
eral blood T cells expressing increased class II
antigens and interleukin 2 (IL-2) receptors are
present in asthmatics admitted into the hospital
for acute flairups (61, 62). In airway allergen
challenge studies of asthmatics, CD4+ T cells
can be demonstrated in BALE at 24 hours after
the challenge associating T cells and LPRs (63).
In allergy skin test studies during the LPR, infil-
trating T cells can be demonstrated expressing
either the IL-2 receptor or class II molecules (64).

A series of important observations made by
Mossman and Coffman in murine T-helper-cell
model systems have direct implications for aller-
gic asthma (65). T helper (Th) cells can be divided
by their cytokine profile into a Th^ type that pro-
duces predominantly IL-2, gamma interferon,
and lymphotoxin, and a Thg type that produces
predominantly IL-4, IL-5, IL-6, and IL-10 (66, 67).
Thj T helper cells produce cellular responses
while Tha T helper cells produce humoral re-
sponses.

It appears, then, that the type of immune re-
sponse that develops to a particular allergen de-
pends on the cytokine profile of the T helper cells.
An allergic asthmatic will generate Th2 helper T
cells to a series of aeroallergens, whereas a
nonatopic individual will develop a Th^ response.
Investigators have isolated T-cell clones specific
for allergen (dust mite) from allergic asthmatics
that produce IL-4; similar clones from nonatopic
nonasthmatic individuals generate gamma inter-
feron in response to the same allergen (68, 69).
These observations validate the existence of Th^
and Thg clones in human allergic diseases and
could have important therapeutic implications.

222

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

Another advance in the understanding of T
cells in allergic asthma was the finding that IL-4
is critical for the synthesis of IgE (70). The im-
portance of IL-4 in IgE production was demon-
strated in co-culture studies in which stimulated
IL-4-producing T cell blasts incubated with donor
B cells from nonallergic individuals supported
IgE synthesis in vitro (71). In these studies, the
ability to maintain IgE synthesis was directly re-
lated to the amount of IL-4 produced by the T cell
blasts. Furthermore, IgE synthesis could be
blocked in culture by adding anti-IL-4 antibody.
In another model system, anti-IL-4 antibody
blunted polyclonal IgE production in response to
parasitic infections, again indicating the impor-
tant role of IL-4 in IgE synthesis (72, 73). The
major biological activities of IL-4 as they relate to
allergic asthma are not confined to the ability to
induce immunoglobulin isotype switch to IgE but
also include activation of other proinflammatory
cells like neutrophils and monocytes (74).

Gamma interferon has a biologic effect oppo-
site to that of IL-4; it inhibits IgE synthesis by B
cells (75). In complementary studies, T cell blasts
were stimulated and co-cultured with donor B
cells, and gamma interferon levels were mea-
sured in the culture supernatants (76). IgE levels
were measured and the opposite effect of IL-4 was
observed. The ability to produce IgE by the donor
B cells was inversely proportional to the gamma
interferon produced by the T cells blasts. In-
creased gamma interferon production by down-
regulating IgE synthesis may have an important
role in regulating the generation of IgE antibod-
ies. The major activity of gamma interferon as it
relates to allergic asthma and the generation of
IgE antibodies is to inhibit IL-4 mediated events,
especially the IgE isotype switch (77). Gamma in-
terferon has been used to treat patients with the
hyper IgE syndrome, a disorder characterized by
recurrent skin infections and markedly elevated
levels of IgE (78). Gamma interferon therapy dra-
matically reduced the IgE levels in these patients
without affecting IgG, IgA, or IgM levels. This
observation has led to several ongoing clinical tri-
als with gamma interferon to treat asthma.

Competition between Thi and Thg clones
may lead to the development of allergic asthma
(79). The atopic individual develops Th2 cells in
response to aeroallergens which produce IL-4 and
little gamma interferon. Th2 cells also produce
another cytokine, IL-10, that can inhibit gamma
interferon production from Thj cells (80). Interac-
tion between the two different types of T helper
clones will lead to either specific IgE antibodies
produced against aeroallergens or a cellular re-

sponse with no IgE antibodies. Potential novel
strategies to treat asthma could include inhibi-
tors of IL-4 (IL-4 receptor antagonists) or inhibi-
tors of IL-10 which block gamma interferon pro-
duction and similarly would inhibit IgE
synthesis. The ability to manipulate this system
will undoubedly lead to better and more specific
asthma therapy.

Other Thg cytokines important in allergic
asthma include interleukin 3 (IL-3) and interleu-
kin 5 (IL-5). IL-5 stimulates production and acti-
vation of eosinophils (81), which can damage
bronchial epithelial cells. IL-5 mRNA has been
demonstrated in T cells from bronchoalveolar
fluid in the late phase reaction after allergen
challenge (82). The local production of IL-5 serves
to activate and recruit eosinophils into the air-
ways, similar to the action of platelet activating
factor and the leukotriene and cyclooxygenase
products. The proliferation and growth of MCx
cells, which are T-cell dependent, is believed to be
mediated through the increased local production
of IL-3 (83, 84) in response to allergen.

Macrophage and Mast Cell
Production of Cytokines

Other cytokine-producing cells have also
been implicated in allergic asthma. Alveolar mac-
rophages produce a wide variety of proinflamma-
tory mediators in asthma (85-87). Recent studies
have also demonstrated that alveolar macro-
phages can become activated via crosslinking of a
low affinity surface IgE receptor (FcE RII) dis-
tinct from the IgE receptor present on mast cells
(FcERI) which results in the production of IL-I-B
and TNF-alpha (88). This may be important clin-
ically since increased levels of IL-1 have been ob-
served in the bronchoalveolar lavage fluid of noc-
turnal asthmatics (89). We have similar data for
activated monocytes-macrophage products in al-
lergic asthma. Our laboratory has recently de-
scribed a novel monocyte-macrophage— derived
mucus secretagog (MMS-68) that causes mucus-
like glycoconjugate release from cultured respira-
tory epithelial cells (90). We have demonstrated
increased spontaneous production of MMS-68
from monocyte culture supernatants in steroid-
dependent asthmatics, again suggesting a role for
mediators produced by alveolar macrophages in
allergic inflammation (91).

Proinflammatory cytokine production is not
restricted to T cells and monocytes. Mast cells, in
addition to releasing histamine, synthesizing a
variety of leukotriene and prostaglandin prod-
ucts, and producing chemotaxic factors for neu-

Vol. 60 No. 3

ASTHMA AS INFLAMMATORY— SPERBER

223

trophils and eosinophils, also synthesize cyto-
kines after the engagement of the FcERI (92).
mRNA for IL-4, IL-5, and IL-6 can be demon-
strated in mast cells after IgE crosslinking (93).
The cytokine profile of these mast cells is identi-
cal to the Thg cell.

Inflammation and
Bronchial Hyperreactivity

If such a large variety of proinflammatory
cells — mast cells, T cells, macrophages, the
eosinophils and neutrophils — cause inflammation
in the airway in allergic asthma through the pro-
duction of mediators, what are the mechanisms
involved that lead to bronchoconstriction? Many
studies have demonstrated that patients never re-
late allergen exposure to a clinical flareup of dis-
ease. Viral infections or other nonspecific stimuli
like exercise and cold air exposure are the major
triggers for asthma (94, 95), not allergen expo-
sure. In order to reconcile this apparent contra-
diction, Cockcroft et al. (96) demonstrated that
after allergen challenge, an allergic asthmatic
has an increase in nonspecific bronchial hyperre-
activity determined by histamine challenge. This
means that after allergen exposure, an allergic
asthmatic required a lower dose of histamine
(PD20) to produce a 20% fall in the forced expira-
tory volume (FEVl) than after preallergen expo-
sure. This is the first demonstration that allergic
airway inflammation can lead to increased non-
specific bronchial hyperreactivity. Seasonal or pe-
rennial allergen exposure in the asthmatic pa-
tient leads to chronic inflammation in the airway,
causing increased bronchohyperreactivity (97).
The increased hyperreactivity makes the patient
more susceptible to nonspecific asthma triggers
like viral infections and exercise (98). The pres-
ence of these cells causes increases in airway hy-
perresponsiveness that can lead to variable air-
way obstruction and to the symptoms of asthma
(99).

Conclusion

If asthma is a chronic inflammatory disease,
antiasthma therapy should consist of antiinflam-
matory medications such as inhaled steroids, not
such agents as theophyline and beta agonists,
which have no antiinflammatory properties.
Many studies have shown that prolonged therapy
with inhaled steroids reduces bronchial hyperre-
activity (100, 101). Furthermore, Adelroth et al.
have shown on bronchial biopsy that treating al-
lergic asthmatics with inhaled steroids leads to
reduction in the numbers of eosinophils, neutro-

phils, and T cells in the airways (102). Further
elucidation of both the cellular and molecular
events involved in the inflammatory processes in
asthma will no doubt lead to more specific and
less toxic therapy.

References

1. Busse WW. The role of inflammation in asthma: a new

focus. Am Rev Respir Dis 1989; 10:72-80.

2. Djukanov R, Roche WR, Wilson JW, et al. Mucosal in-

flammation in asthma. Am Rev Respir Dis 1990; 142:
434^57.

3. Laitinen LA, Heino M, Laitinen M, et al. Damage of the

airway epithelium and bronchial reactivity in mild
asthma and after bronchial reactivity in patients with
asthma. Am Rev Respir Dis 1985; 131:599-606.

4. Djukanovic R, Wilson J, Britten K, et al. Quantition of

mast cells and eosinophils in the bronchial mucosa of
symptomatic atopic asthmatics and healthy control
subjects using histochemistry. Am Rev Respir Dis
1990; 142:863-871.

5. Wardlaw AJ, Dunnett S, Gleich GJ, et al. Eosinophils

and mast cells in bronchoalveolar lavage in subjects
with mild asthma. Am Rev Respir Dis 1988; 137:62-
69.

6. Gibson PG, Girgis-Gabardo A, Morris MM, et al. Cellu-

lar characteristics of sputum from patients with
asthma and chronic bronchitics. Thorax 1989; 44:689—
692.

7. Beasley, Roche WR, Roberts JA, Holgate ST. Mucosal

inflammation in asthma. Am Rev Respir Dis 1989;
139:808-817.

8. Metzer WJ, Zavala D, Richarson HB, et al. Local aller-

gen challenge and bronchial alveolar lavage of allergic
asthmatic lungs. Am Rev Respir Dis 1987; 135:433-
440.

9. O'Byrne P, Dolovitch J, Heagreave FE. Late asthmatic

responses. Am Rev Respir Dis 1987; 134:740-751.

10. Baushey HA, Holtzman MJ, Sheller JR, Nadel JA. Bron-

chial hyperreactivity. Am Rev Respir Dis 1985; 121:
389-414.

11. Holgate ST, Beasley R, Twentyman OP. The pathogen-

esis and significance of bronchial hyperresponsiveness
in airways disease. Clin Sci 1987; 73:501-572.

12. Bleecker ER. Airways hyperresponsiveness and asthma:

significance and treatment. J Allergy Clin Immunol
1985; 75:21-24.

13. Gergen PJ, Turkeltaub PC. The association of allergen

skin test reactivity and respiratory disease among
whites in the U.S. population. Arch Intern Med 1991;
151:487-194.

14. Kalliel JN, Goldstein BM, Broman SS, et al. High fre-

quency of atopic asthma in a pulmonary chest clinic
population. Chest 1989; 96:1336-1340.

15. Licorish K, Hovey HS, Kozak P, et al. Role of Alternia

and Penicillium spores in the pathogenesis of asthma.
J Allergy Clin Immunol 1985; 76:819-826.

16. Burrows B, Martinez FD, Halonen M, et al. Association

of asthma with serum IgE levels and skin test reactiv-
ities to allergens. N Engl J Med 1989; 320:271-277.

17. PoUart SM, Chapman M, Fiooco GP, et al. Epidemiology

of acute asthma: IgE antibodies as a risk factor for
emergency room visits. J Allergy Clin Immunol 1989;
83:875-882.

18. Reid MJ, Moss RB, Hsu YP, et al. Seasonal asthma in

224

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

northern California: allergic causes and efficiency of
immunotherapy. J Allergy Clin Immunol 1986; 78:
590-600.

19. Sperber K, Kiendler D, Yu LM, Nayak H, Pizzimenti A.

Prevalence of atopy in an inner-city asthmatic popu-
lation. Mt Sinai J Med 1993; 60:227-231.

20. Galli SJ, Austen FK, Mast cell and basophil differentia-

tion in health and disease. New York: Raven Press,
1989.

21. Galli SJ, Lichtenstein LM. Biology of mast cells and ba-

sophils In: Middleton E, ed. Allergy principles and
practice. St. Louis: C.V. Mosby Company, 1988.

22. Bradley BL, Azzawi M, Jacobson M, et al. Eosinophils,

T-lymphocytes, mast cells, neutrophils, and macro-
phages in bronchial biopsies from atopic asthmatics:
comparison with atopic non-asthma and normal con-
trols and relationship to bronchial hyperresponsive-
ness. J Allergy Clin Immunol 1991; 138:661-671.

23. Irani AA, Scwartz LB. Neutral proteases as indicators of

human mast cell heterogenity. In: Schwartz LB, ed.
Neutral protease of mast cells. Basel: Karger, 1990.

24. Bienstock J, Befus AD, Denburg JA. Mast cell heteroge-

neity: basic questions and clinical implications. In: Be-
fus AD, Bienstock J, Denburg JA, ed. Mast cell differ-
entiation and heterogenity. New York: Raven Press,
1986.

25. Irani AM, Craig SS, DeBlois G, et al. Deficiency of the

tryptase-positive, chymase-negative mast cell type in
gastrointestinal mucosa of patients with defective
T-lymphocyte function. J Immunol 1987; 138:4381-
4390.

26. Galli SJ, Dvorak AM, Dvorak HF. Basophils and mast

cells: morphologic insights into their biology, secretory
patterns and function. Prog Allergy 1984; 34:1-24.

27. Benyon RC, Lowman MA, Church MK. Human skin

mast cells: their dispersion, purification and secretory
characteristics. J Immunol 1987; 138:861-880.

28. Kitamura Y, Go S, Hatahata T, et al. Decrease of mast

cells in WAV mice and their increase by bone marrow
transplantation. Blood 1978; 52:447^57.

29. Galli SJ. New approaches for the analysis of mast mat-

uration, heterogeneity and function. Fed Proc 1987;
46:1906-1922.

30. Stevens RL, Rothenberg ME, Levi-Schaffer F, Austen

KF. Ontogeny of in vitro-differentiated mouse mast
cells. Fed Proc 1987; 46:1915-1930.

31. Dvorak AM, Newball HH, Dvorak HF, Lichenstein LM.

Antigen induced IgE-mediated degranulation of hu-
man basophils. Lab Invest 1980; 43:126-135.

32. Baushey HA, Holtz MJ, Sheller JR, Nadel JA. Bronchial

hyperreactivity. Am Rev Respir Dis 1980; 121:389-
414.

33. Lichtenstein LM. In: Middleton E, ed. Allergy principles

and practice. Chap. 12. St. Louis: C.V. Mosby, 1988.

34. Dvorak AM, Galli SJ, Morgan E, et al. Anaphylactic de-

granulation of guinea pig basophilic leukocytes. I: Fu-
sion of granule membranes and cytoplasmic vesicles:
formation and resolution of degranulation sacs. Lab
Invest 1981; 44:174-182.

35. Nelson HS. Diagnostic procedures in allergy: I: Allergy

skin testing. Ann Allergy 1983; 51:411-425.

36. Dunsky EH, Zweiman B. The direct demonstration of

histamine release in allergic patients in the skin using
a skin chamber technique. J Allergy Clin Immunol
1978; 62:127-138.

37. Atkins P, Green GR, Zweiman B. Histologic studies of

human skin test responses to ragweed, compound 48/

80, and histamine. J Allergy Clin Immunol 1973; 51:
263-272.

38. Davies P, Baily PJ, Goldenberg MM. The role of arachi-

donic acid oxygenation production in pain and inflam-
mation. Crit Rev Immunol 1984; 31;307-327.

39. Burrall BA, Payan DG, Goetzl D. Arachidonic acid-de-

rived mediators of hypersensitivity and inflammation.
In: Middleton E, ed. Allergy principles and practice.
St. Louis: C.V. Mosby, 1988.

40. Platshon L, Kaliner M. The effect of the immunologic

release of histamine upon human cyclic nucleotide lev-
els and prostaglandin generation. J Allergy Clin Im-
munol 1978; 62:1113-1121.

41. Spannhake EW, Hysman AL, Kadowitz PJ. Bronchoac-

tive metabolites of arachidonic acid and their role in
airway function. Prostaglandins 1981; 22:1013-1026.

42. Steveson DD, Mathison DA. Aspirin sensitivity in asth-

matics: when may this drug be safe? Postgrad Med
1985; 78:111-121.

43. Ford-Hutchinson AW. Leukotriene involvement in

pathologic processes. J Allergy Clin Immunol 1984;
74:437^51.

44. Dahlen S, Hansson G, Hedquist R, et al. Allergen chal-

lenge of lung tissue from asthmatics elicts bronchial
contraction that correlates with the release of leuko-
trienes C4, D4 and E4. Proc Nat Acad Sci USA 1983;
80:1712-1720.

45. Goetzl EJ, Scott WA. Proceedings of a conference on reg-

ulation of cellular activities by leukotrienes and other
lipoxygenase products of arachidonic acid. J Allergy
Clin Immunol 1984; 74:309-359.

46. Isreal E, Dermarkian R, Rosenberg M, et al. The effect of

5-lipooxygenase inhibitor on asthma induced by cold
air. N Engl J Med 1990; 323:1704-1740.

47. Manning PJ, Watson RM, Margolkee DJ, et al. Inhibi-

tion of exercise induced asthma by MK-571, a potent
leukotriene receptor antagonist. N Engl J Med 1990;
323:1736-1739.

48. Lynch JM, Worthen GS, Henson PM. Platelet activating

factor. In: Burkle DR, Smith H, eds. Development of
anti-asthma drugs. London: Butterworth, 1984.

49. Pinchard RN, McManus LM, Hamalan DJ. Chemistry

and biology of PAF. In: Weisman G, ed. Advances in
inflammation. Vol. 4. New York: Raven Press, 1984.

50. Lynch JM, Henson PM. The intracellular retention of

newly synthesized platelet activating factor. J Immu-
nol 1986; 137:2653-2660.

51. Cuss FM, Dixon CMS, Barnes PJ. Effects of inhaled

platelet activating factor on pulmonary function and
bronchial responsiveness in man. Lancet 1986; 2:189-
196.

52. Majorad M, Said SI. Platelet activating factor increases

microvascular permeability. Am Rev Respir Dis 1982;
125:276-288.

53. Ingraham LM, Coates TD, Allen JM, et al. Metabolic

membrane and functional responses of human poly-
morphonuclear leukocytes to platelet activating fac-
tor. Blood 1982; 59:1259-1273.

54. Worthen GS, Gaines Al, Mitchel BC, et al. Platelet ac-

tivating factor causes neutrophil accumulation and
edema in rabbit lungs. Chest 1983; 83:13S-20S.

55. Slifman NR, Adolphson CR, Gleich GJ. Eosinophils: bio-

chemical and cellular aspects. In: Middleton E, ed. Al-
lergy principles and practice. St Louis: C.V. Mosby,
1988.

56. Cockcroft DW. Mechanisms of perennial allergic asthma.

Lancet 1983; 2:253-256.

Vol. 60 No. 3 ASTHMA AS INFLAMMATORY— SPERBER 225

57. Barnes PJ. A new approach to the treatment of asthma.

N Engl J Med 1984; 132:1517-1527.

58. Cockcroft DW. Airway hyperresponsiveness: therapeutic

implication. Ann Allergy 1987; 59:405^14.

59. Frew AJ, Kay AB. The relationship between infiltrating

CD4+ lymphocytes, activated eosinophils and the
magnitude of the allergen induced late phase cutane-
ous reaction in man. J Immunol 1988; 85:6880-6884.

60. Pene J, Rousset F, Briere F, et al. IgE production by

normal lymphocytes is induced by interleukin 4 and
suppressed by gamma interferon, alpha and PGE2.
Proc Natl Acad Sci USA 1988; 85:6880-6884.

61. Brown PH, Crompton GK, Grening AP. Proinflamma-

tory cytokines in acute asthma. Lancet 1991; 338:590-
503.

62. Corrigan CJ, Kay AB. CD4 lymphocyte activation in

acute severe asthma. Am Rev Respir Dis 1990; 140:
970-977.

63. Robinson DS, Bently AM, Durham SR, Kay AB. T helper

lymphocyte activation in BAL from atopic asthmatics:
correlation with eosinophil, symptoms, and bronchial
hyperresponsiveness. Presented at 3rd International
Conference, European Respiratory Review, Vienna,
Austria, Jun 20-22, 1991. Abstract.

64. Kay AB, Ying S, Varney, et al. Expression of the cyto-

kine cluster interleukin-3 (IL-3), IL-4, IL-5 and gran-
ulocyte/macrophage colony stimulating factor in aller-
gen induced late phase cutanous responses in atopic
subjects. J Exp Med 1991; 172:775-778.

65. Mossman TR, Coffman RL. THl and TH2 cells: different

patterns of lymphokine secretion lead to different
functional properties. Ann Rev Immunol 1989; 7:145-
165.

66. Wierenga EA, Snoek M, Bos JD, et al. Comparison of

diversity and function of house dust mite-specific T
lymphocytes from atopic and non-atopic donors. Eur J
Immunol 1990; 20:1519-1523.

67. Mossmamm TR, Coffman RL. Two types of mouse helper

T cell clones: implications for immune regulation. Im-
munol Today 1987; 8:233-227.

68. Wierenga EA, Snoek M, de Groot C, et al. Evidence for

compartmentalization of functional subsets of CD4 +
T lymphocytes in atopic patients. J Immunol 1990;
144:4651-4660.

69. Parronchi P, Macchia D, Piccini MP, et al. Allergen and

bacterial antigen-specific T cell clones established
from atopic donors show a different profile of cytokine
production. Proc Natl Acad Sci USA 1991; 188:4538-
4542.

70. Yokata Y, Otsuka T, Mossman T, et al. Isolation and

cloning of a human cDNA clone homologous to mouse
BSF-1 that expresses B cell/T cell stimulating activity.
Proc Natl Acad Sci USA 1986; 83:5894-5900.

71. Del Prete GF, Maggie E, Parronchi P, et al. IL-4 is an

essential factor for IgE synthesis induced in vitro by
human T cell clones and their supernatants. J Immu-
nol 1988; 140:4193^199.

72. Finkelman F, Katona IM, Urban JF, et al. Suppression

of in vivo polyclonal IgE responses by monoclonal an-
tibody to lymphokine BSF-1. Proc Natl Acad Sci USA
1986; 83:9675-9685.

73. Street NE, Schumache JH, Fony FA, et al. Heterogene-

ity of mouse T cells: evidence from bulk cultures and
limiting dilutions for precursors of TH-1 and TH-2
cells. J Immunol 1990; 144:1629-1639.

74. Vercelli D, Jabara HH, Lee B, et al. Human recombinant

interleukin 4 induced FcER2/CD-23 on normal mono-
cytes. J Exp Med 1988; 167:1406-1416.

75. Romagnani S. Regulation and deregulation of human

IgE synthesis. Immunol Today 1990; 11:316-324.

76. Coffman RL, Carty J. A T-cell activity that enhances

polyclonal IgE production and its inhibition by gamma
interferon. J Immunol 1986; 136:949-961.

77. Finkelman FD, Holmes J, Kotona IM, et al. Lymphokine

control of in vivo immunoglobulin isotype selection.
Am Rev Immunol 1990; 8:303-334.

78. King CL, Gallin JI, Malech HL, et al. Regulation of im-

munoglobulin production in the hyper-immunoglobu-
lin syndrome recurrent infection syndrome. Proc Natl
Acad Sci USA 1989; 86:10085-10091.

79. Gojeniski TF, Schnell SR, Fitch FW. Evidence implicat-

ing utilization of different T cell receptor-associated
signaling pathways by TH-1 and TH-2 clones. J Im-
munol 1990; 144:4110-4120.

80. Moore KW, Viera P, Fiorentino DF, et al. Homology of

cytokine synthesis inhibitory factor (IL-10) to the Ep-
stein-Barr virus gene BCRFl. Science 1990; 2348:
1230-1244.

81. Lopez AF, Sanderson CJ, Gamble JR, et al. Recombinant

human interleukin-5 is a selective activation of hu-
man eosinophil function. J Exp Med 1988; 219-224.

82. Robinson DS, Hamid Q, Ying S, et al. Predominant TH2-

like bronchoalveolar T-lymphocyte populations in
atopic asthma. N Engl J Med 1992; 326:298-304.

83. Kirshenbaum AS, Goff JP, Dreskin SC, et al. IL-3 de-

pendent growth of basophil-like cells and mast cells
from human bone marrow. J Immunol 1989;
142:2427-24.

84. Anderson DM, Lyman SP, Baird A, Wigmal JM. Molec-

ular cloning of mast cell growth factor, a hematopoi-
etin that is active in both membrane bound and solu-
ble forms. Cell 1990; 63:235-245.

85. Damon M, Chavis C, Crattes de Paulet M, et al. Arachi-

donic acid metabolism in alveolar macrophages: a
comparison of cells from healthy subjects, allergic
asthmatics, and chronic bronchitis patients. Pros-
taglandins 1987; 34:291-309.

86. Wilkerson JRW, Crea A, Clark TJH, Lee TH. Identifica-

tion and characterization of a monocyte-derived factor
in corticosteroid resistant bronchial asthma. J Clin In-
vest 1989; 84:1930-1941.

87. Gosset P, Tonnel AB, Joseph M, et al. Secretion of a

chemotactic factor for neutrophils and eosinophils
from asthmatic patients. J Allergy Clin Immunol
1984; 74:827-834.

88. Borish L, Mascali J, Rosenwasser LJ. IgE-dependent cy-

tokine production by human peripheral blood mono-
nuclear phagocytes. J Immunol 1991; 146:93-101.

89. Martin R, Borish L, Mascali J, Rosenwasser. Interleukin

1-B production by alveolar macrophages derived from
nocturnal asthmatics. J Allergy Clin Immunol 1991;
87:215.

90. Gollub E, Goswami S, Sperber K, Marom Z. Isolation and

characterization of a macrophage derived high mole-
cule weight protein involved in the regulation of mu-
cus-like glycoconjugate secretion. 1992; 89:656-664.

91. Sperber K, Gollub Goswami S, Kalb T, Mayer L, Marom

Z. In vivo detection of a novel macrophage protein in-
volved in the regulation of mucus-like glycoconjugate
secretion. Am Rev Respir Dis 1992; 1589-1597.

92. Plaut M, Pierce JH, Watson CJ, et al. Mast cell lines

produce lymphokines in response to crosslinkage of
FcERI or to calcium ionophores. Nature 1989; 339:64.

226

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

93. Burd PR, Roders HW, Gordon JR, et al. Interleukin 3

dependent and independent mast cells stimulated with
IgE and antigen express multiple cytokines. J Exp
Med 1989; 170:245-255.

94. Reed CE. Review of therapeutic modulation in asthma.

In: Reed C, ed. Proceedings of the XII International
Congress of Allergy and Clinical Immunology. St.
Louis: C.V. Mosby, 1983: 336-373.

95. Murray AB, Fergusen AC, Morrison BJ. Diagnosis of

house dust mite allergy in asthmatic children: what
constitutes a positive history? J Allergy Clin Immunol
1983; 78:21-29.

96. Cockcroft DW, RufFm RE, Dolovich J, Hergreave FE. Al-

lergen induced increased in non-allergic bronchial re-
activity. Clin Allergy 1977; 7:503-513.

97. Zimmerman B, Feanny L, Reisman J, et al. Allergy in

asthma: I. The dose relationship of allergy to severity
of childhood asthma. J Allergy Clin Immunol 1988;
81:63-70.

98. Chung KF. Role of inflammation in the hyperreactivity

of the airways in asthma. Thorax 1986; 41;657-662.

99. James AL, Pare PD, Hogg JC. The mechanisms of air-

way narrowing in asthma. Am Rev Respir Dis 1989;
139:242-246.

100. Cockroft DW, Murdoch KY. Comparative effects of in-

haled salbutamol, sodium chromoglycate and beclo-
methamesone on allergen induced early asthmatic re-
sponses and increased bronchial responsiveness to his-
tamine. J Allergy Clin Immunol 1987; 79:734-740.

101. Dahl R, Johanson SA. Importance of duration of treat-

ment with inhaled budesonide on the immediate and
late bronchial reaction. Eur J Respir Dis (Suppl) 1982;
122:167-175.

102. Adelroth, Rosenhall L, Johansson S, Linden M, Venge P.

Inflammatory cells and eosinophil activity in asthmat-
ics investigated by bronchoalveolar lavage. Am Rev
Respir Dis 1990; 142:91-99.

Prevalence of Atopy in an Inner-City
Asthmatic Population

Kirk Sperber, M.D., David Kendler, M.D., Lai Ming Yu, M.D., Hemal Nayak, M.D., and

Andrew Pizzimenti, B.S.

Abstract

Seventy-three patients at The Mount Sinai Hospital Emergency Room were investigated
to determine the prevalence of atopy in asthma in a predominantly black and Hispanic
inner-city population. Serum IgE levels and radioallergosorbent tests (RASTs) to eight
common inhalant allergens were measured in both the asthmatic group and a nonasth-
matic emergency-room control group. The mean total IgE level for the asthma group was
263.8 lU/mL compared to 63.8 lU/mL in the control group (p = 0.032), and 60% of the
asthmatics had IgE levels in the atopic range OlOO lU/mL). Increases in IgE were asso-
ciated with age under 50 years but did not reach statistical significance. Cockroach, dust
mite, cat, and dog were the most common RASTs in the asthmatic group; there were no
positive RASTs in the control group. There was a correlation ip = 0.04) between age (less
than 50 years) and increased numbers of positive RASTs. These results are similar to
those of other studies that have associated atopy with asthma in rural and suburban
populations. These data demonstrate that atopy is common in the asthmatic patients seen
in The Mount Sinai Hospital Emergency Room and strongly suggest that management of
atopic factors should become routine in the care of adult asthmatic patients.

Asthma is a chronic inflammatory disease with a
multifactorial etiology which affects approxi-
mately 5% of the population of the United States
(1). Recent national studies have revealed that
the incidence and severity of this disease are in-
creasing, especially in inner-city minority popu-
lations (2, 3). Although asthma was once consid-
ered a disease occasionally associated with
allergic factors, several recent large studies have
clearly established a connection between atopy
and asthma in a large percentage of asthmatics
(4—6). These studies of adult patients have come
primarily from rural and suburban populations
and have associated either elevated IgE levels or
radioallergosorbent tests (RASTs) positive for a

From the Divisions of Clinical Immunology (KS, LMY, HN,
AP) and Endocrinology (DK), Department of Medicine, The
Mount Sinai Medical Center, New York City, NY. Address
correspondence and reprint requests to Dr. Kirk Sperber, Di-
vision of Clinical Immunology, Box 1089, One Gustave Levy
Place, New York, NY 10029.

limited number of aeroallergens with emergency
visits for asthma treatment. Burrows et al. (4), in
a large Arizona study, found higher levels in 90%
of patients below the age of 40 years reporting to
local emergency rooms for asthma treatment
than in age-matched, nonasthmatic controls.
Similarly, Pollart et al. in two studies, one in ru-
ral Virginia (5) and one in northern California
(6), associated increased emergency visits for
asthma in adult patients younger than 50 years of
age with RASTs positive for dust mite and rye
grass, respectively.

Although the connection between atopy and
asthma has been established in these rural and
suburban populations, few studies have at-
tempted to identify atopic factors in inner-city
minority populations (7). To determine the prev-
alence of atopy in asthmatics from a predomi-
nantly black and Hispanic inner-city population,
we measured IgE and RAST levels in 73 asth-
matic patients and in 30 age-matched, nonasth-
matic control patients seen in The Mount Sinai
Emergency Room. The objectives of this study

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

227

228

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

TABLE 1

Distribution of Age,

Sex. and Race

of the Study Population

rVbLnind, gruup

(n = 30)

(n = 73)

Age

44.7 ± 20.7

45.9 ± 15.3

Sex

{% female)

64%

65%

{% male)

36%

35%

Race

(Hispanic)

82%

79%

(black)

15%

16%

(white)

Both the control and asthma group were predominantly His-
panic and black. The mean duration of disease was 14.6 years.

were (a) to assess the prevalence of elevated IgE
levels in a population of asthmatics compared to a
nonasthmatic control group and (b) to determine
which allergens are most common in the asth-
matic group.

Methods

Study Population. Patients seen in The
Mount Sinai Hospital asthma room from Febru-
ary 1990 to June 1990 were enrolled in this study
after they had been evaluated by an emergency
room physician and judged to be having acute air-
way obstruction requiring bronchodilator ther-
apy. Patients were enrolled by two of the investi-
gators (HN and LMY). Peak flow measurements
before and after bronchodilator therapy were
made with mini-Wright peak flow meter (Arm-
strong Industries, Northbrook, ID in 46 of the 73
patients with asthma, and each patient demon-
strated marked improvement after treatment.
The rest of the asthmatic patients were too ill or
unwilling to comply before bronchodilator ther-
apy. Serum samples were obtained from all the
asthmatic patients prior to therapy. Demographic

TABLE 2

Distribution of Asthma Medications

No. of patients

Oral steroids + beta agonist

+ theophyline

Beta agonist + theophylline

Beta agonist + theophylline +

inhaled

steroid

Beta agonist + theophylline +

inhaled

steroid + cromolyn

Beta agonist + inhaled steroid

Beta agonist

None

Fifty-two patients, including 22 taking oral prednisone, re-
quired multiple medications to control their disease.

data, patient medications, and duration of disease
were determined from the emergency room
sheets. Serum samples from age-matched controls
were also obtained during the same period from
patients entering the emergency room for evalu-
ation of medical diagnoses other than asthma, in-
cluding angina, pneumonia, trauma, renal colic,
and upper respiratory infection.

Laboratory Assessment. Laboratory mea-
surements for total serum IgE levels were deter-
mined by enzyme immunoassay (PRIST) (Phar-
macia, Piscataway, NJ). RASTs for timothy grass
pollen, elm pollen, ragweed, cat, dog, Aspergillus
fumigatus, house dust mite, and cockroach were
performed utilizing a commercially available kit
from Pharmacia.

Statistical Analysis. Statistical analysis was
performed using the SAS statistical program
package (SAS Institute Inc., Gary, NC). Data
were compared using a two-tailed Student ^-test.
An alpha value of 0.05 was established as the
criterion for statistical significance.

Results

Study Population. The asthmatic population
of 73 was composed of 64% women and 36% men;
the control group of 30 was composed of 70%
women and 30% men. The mean age of the study
population was 45.9 years (range 18-74 years),
and the mean age of the control group was 44.7
years (range 18-67 years). Sixty of the asthmatic
patients were Hispanic, 11 were black, and 2 were
white. There were 40 Hispanic women, 7 black
women, 20 Hispanic men, 4 black men, and 2
white men. Eighteen patients were older than 50
years, and 55 were younger than 50. The popula-
tion used in this study reflects the demographics
of the community that surrounds The Mount
Sinai Hospital (see Table 1).

The severity of asthma in the study group is
reflected in the mean duration of disease, 14.6
years (range 2 months to 30 years), and the
amount of medications required to control the
asthma. Fifty-four of the 73 asthmatics, including
22 who were taking oral prednisone, required
multiple medications to control their disease (Ta-
ble 2).

Incidence of Elevated IgE Levels in Asth-
matic Patients. Figure 1 illustrates the range of
distribution of IgE values in both the asthmatic
and control groups. There is a statistically signif-
icant difference (p = 0.032) between the mean
IgE level for the asthmatic group, 263.84 lU/mL

Vol. 60 No. 3

ATOPY IN INNER CITY— SPERBER ET AL

229

2000

p-0.03

1 000

500

0 -

o 8 o
o Q o o

No Asthma

• t •

iiil

Asthma

Fig. 1. Distribution of IgE levels in asthmatic and control groups. IgE levels are significantly (p
asthmatics.

0.03) increased in the

(range 0-2000 lU/mL) and the control group, 63.3
lU/mL. There were no statistical differences in
IgE levels either between men and women or be-
tween blacks and Hispanics in the asthmatic or
control groups. Fifty of the 73 asthmatic patients
had IgE levels greater than 100 lU/mL, which is
defined as the lower limit of the atopic range.
There were no statistical differences in IgE levels
between those patients who were older than 50
years of age and those younger than 50. IgE levels

TABLE 3

Percentage of Positive RASTs in the Group of 73 Patients
with Asthma

Positive RASTS

Allergen (%)

Dust mite 24

Dog 28

Cat 38

Cockroach 68

Ragweed 17

Grass 15

Tree 14

Aspergillus fumigatus 7

Cockroach, cat, dog, and dust mite were the most common
allergens. The mean number of positive RASTs per patient
was 2.5.

of patients who were taking oral prednisone were
lower than those not taking oral steroids, but this
did not reach statistical significance (data not
shown).

Prevalence of Positive RASTs among
Asthma Patients. Table 3 demonstrates the dis-
tribution of positive RASTs in the asthmatic pa-
tients. Cockroach was the most common allergen
in our study, followed by cat, dog, dust mite, rag-
weed, trees, grass, and Aspergillus fumigatus.
There were no positive RAST results in the con-
trol group. There was a correlation (p = 0.04)
between the number of positive RASTs and age
less than 50 years (Fig. 2).

Discussion

The incidence and severity of asthma, espe-
cially in inner-city minority populations, has in-
creased markedly in the past ten years for as yet
undetermined reasons (8). Although inadequate
asthma treatment and lack of access to proper
medical facilities may be influencing this in-
crease in mortality and morbidity, other disease-
specific factors may be playing a role as well.
Over the past ten years it has become apparent
that increases in bronchial airway hyperactivity

230

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

1 5

Fig. 2. A statistically significant difference (p = 0.04) was
demonstrated between the mean number of positive RASTs in
asthmatics younger than 50 years of age and in asthmatics
older than 50 years.

associated with allergen exposure contribute to
both the pathogenesis and severity of asthma (9,
10).

The asthmatic population treated in The
Mount Sinai Emergency Room is almost entirely
black and Hispanic and predominantly female
(64%) (Table 1). Although there is no known sex
predisposition for asthma, the predominance of
Hispanic and black women observed in our study
is similar to the results of Karetzky's study in the
South Bronx in 1977 (7). It is uncertain whether
this increase in female asthmatics reflects in-
creased severity of disease in black and Hispanic
women or increased utilization of emergency
room services by this group.

Our asthmatic population appears to have
relatively severe disease, as reflected in the
amount of medication required to control the
asthma, including 22 patients taking corticoste-
roids (Table 2). Increased severity of asthma is
well recognized in inner-city black populations (8)
but is less well described in Hispanic groups.

Our data demonstrate that atopy, defined as
an elevated IgE level (>100 lU/ml), is extremely
common (60% of patients) among asthmatics seen
in The Mount Sinai Hospital asthma room. This
increase in both total IgE and specific IgE di-
rected against aeroallergens (RAST) appears to
be disease specific since it was not found in the
age-matched control group. Although IgE levels
were higher in the asthmatics younger than 50
years (4-6), this difference did not reach statisti-
cal significance. There was, however, a statisti-
cally significant association ip = 0.04) between
the number of positive RASTs and age (under 50
years) demonstrating that specific antiallergen
IgE antibodies are more common in younger asth-

matics, which is consistent with previous studies
(4—6). There was no correlation between either
the duration of asthma or the medications needed
to control disease and the IgE level or the number
of positive RASTs (data not shown). Although
60% of the patients were found to have elevated
IgE levels and positive RASTs, no conclusions re-
garding the role of acute allergen exposure and
the asthma that brought the patients to the emer-
gency room can be made. Repeated studies (11-
13) have demonstrated that patients generally do
not associate allergen exposure with an asthma
attack. However, extensive evidence demon-
strates that prolonged allergen exposure in-
creases nonspecific bronchial reactivity through
the accumulation of inflammatory cells. Neutro-
phils and eosinophils in the airway may lower the
threshold for nonspecific triggers, such as viral
infections and exercise, that can provoke acute
bronchoconstriction (14-16). The atopic nature of
the asthmatic population is similar to other stud-
ies performed in rural and suburban populations
(4-6).

The asthmatic patients were allergic to an
indoor panel of allergens, including cockroach,
cat, dog, and dust mite (Table 3). Cockroach was
the most common allergen in the asthmatic pa-
tients and has been recognized in numerous stud-
ies as a major allergen in both allergic rhinitis
and asthma (17-19). Cockroach allergens are well
characterized (20) and have been demonstrated
on bronchoprovocation to induce both immediate
and late-phase reactions (20). Although skin-test
positivity to cockroach has also been reported by
several investigators in diverse groups of asth-
matics (21-23), it is especially common among in-
digent populations of asthmatics usually living in
public housing (24). Cockroach and other indoor
allergens, especially dust mite, are a perennial
source of antigen-inducing airway inflammation.
Constant exposure to these allergens may cer-
tainly contribute to airway inflammation in this
patient population.

In conclusion, elevations in total IgE and spe-
cific IgE antibodies directed against an indoor
panel of allergens, including cockroach, cat, dog,
and dust mite, are present in 60% of the asthmat-
ics seen in The Mount Sinai Hospital asthma
room. These allergens are a perennial source of
airway inflammation in this population. The
atopic component of asthma in these patients
must be taken into account in the treatment of
this disease, and attempts to reduce airway in-
flammation by limiting allergen exposure must
be utilized.

Vol. 60 No. 3

ATOPY IN INNER CITY— SPERBER ET AL

231

Acknowledgements

We thank Dr. Barbara Richardson and medical housestaff for
their help in recruiting the patients in the study and Sonia
Makuku for performing the IgE levels and the RASTs.

References

1. National Center for Health Statistics, DHHS publication

1984, No. (PHS) 85-1232. Washington DC: US Govern-
ment Printing Office.

2. Death due to chronic obstructive disease and allied con-

ditions. MMWR 1984; 35:507-510,

3. Mak H, Johnston P, Abbey H, Talamo RC. Prevalence of

asthma and health service utilization of asthmatic chil-
dren in an inner city. J Allergy Clin Immunol 1982;
70:367-372.

4. Burrows B, Martinez FD, Halonen M, et al. Association of

asthma with serum IgE levels and skin test reactivity
to allergens. N Engl J Med 1989; 320:271-277.

5. Pollart SM, Chapman MD, Fiocco GP, et al. Epidemiology

of acute asthma: IgE antibodies to common inhalant as
a risk factor for ER visits. J Allergy Clin Immunol
1989; 83:875-882.

6. Pollart SM, Reid MJ, Fling JA, et al. Epidemiology of ER

asthma in Northern California: association with IgE
antibody to rye grass pollen. J Allergy Clin Immunol
1988; 82:224-230.

7. Karetzky M. Asthma in the South Bronx: clinical and ep-

idemiological characteristics. J Allergy Clin Immunol
1977; 60(6):383-390.

8. Evans R, Mullally DI, Wilson RW, et al. National trends

in the morbidity and mortality of asthma in the United
States: prevalence, hospitalization and death from
asthma over two decades: 1965-85. Chest 1987;
91(Suppl):65S-74S.

9. Cockcroft DW. Mechanism of perennial allergic asthma.

Lancet 1983; 2:253-256.

10. Chapman MD, Pollart SM, Luczynska CM, et al. Hidden

allergic factors in the etiology of asthma. Chest 1988;
94(11:185-190.

11. Stevenson DD, Mathison DA, Ian EM, Vaughan JH. Pro-

voking factors in bronchial asthma. Arch Intern Med
1975; 135:777-783.

12. Reed CE. Review of therapeutic modalities in asthma, In:

Reed CE, ed. Proceedings of the XII International Con-
gress of Allergy and Clinical Immunology. Washington
DC: CV Mosby, 336-373.

13. Murray AB, Ferguson AC, Morrison BJ. Diagnosis of

house dust mite allergy in asthmatic children: what
constitutes a positive history? J Allergy Clin Immunol
1983; 78:21-29.

14. Holgate ST, Beasley R, Twentymen OP. The pathogenesis

and significance of bronchial hyperresponsiveness in
airways disease. Clin Sci 1987; 73:651-672.

15. Barnes PJ. A new approach to the treatment of asthma. N

Engl J Med 1989; 321(22):1517-1527.

16. Zimmerman B, Feanny S, Reisman J, et al. Allergy in

asthma. I. The dose relationship of allergy to severity of
childhood asthma. J Allergy Clin Immunol 1988; 81:
63-70.

17. Ohman J. Allergen immunotherapy in asthma: evidence

for efficacy. J Allergy Clin Immunol 1989; 84(2): 133-
140.

18. Kang B. Study on cockroach antigen as probable caus-

ative agent in bronchial asthma. J Allergy Clin Immu-
nol 1976; 58:365-371.

19. Shulaner FA. Sensitivity to the cockroach in three groups

of allergic children. Pediatrics 1970; 45:465^70.

20. Bernton HS, Brown H. Cockroach allergy. II. The relation

of infestation to sensitization. South Med J 1967; 60:
852-857.

21. Lan JL, Lee DT, Wu CH, et al. Cockroach hypersensitiv-

ity: preliminary study of allergic cockroach asthma in
Taiwan. J Allergy Clin Immunol 1988; 82:736-740.

22. Kang B, Sulit N. A comparative study of prevalence of

skin hypersensitivity to cockroach and house dust an-
tigens. Ann Allergy 1978; 41:333-342.

23. Kang B, Vellody D, Homberger H, Yuniniger JW. Cock-

roach causes allergic asthma; its specificity and immu-
nologic profile. J Allergy Clin Immunol 1979; 63:80-88.

24. Bernton HS, McMahon TF, Brown H. Cockroach asthma.

Br J Dis Chest 1972; 66:61-66.

Submitted for publication August 1991.
Revision received November 1991.

Oligoclonal Banding in AIDS
and Hemophilia

Etta B. Frankel, M.D., Michael L. Greenberg, M.D., Sonya Makuku, B.A., M.T. (ASCP), and

Shaul Kochwa, Ph.D.

Abstract

Hypergammaglobulinemia is a consistent finding in patients within the AIDS spectrum
and with hemophilia A. Serum samples fi"om patients with these conditions were analyzed
for the presence of oligoclonal banding, using a high-resolution serum protein electropho-
resis system. The incidence of banding is significantly greater in well homosexuals who
are HIV-antibody positive and in patients with pre-AIDS-related complex, AIDS-related
complex, AIDS with opportunistic infections, and AIDS with Kaposi's sarcoma than in
normal blood donors. The incidence of banding is similar to controls in patients with
hemophilia who have received either no blood products, cryoprecipitate only, or limited
infusions of factor VIII concentrate. In patients who have received frequent infusions of
factor VIII concentrate, the incidence of banding significantly increases. Thirteen of sixty-
seven hemophiliac patients developed AIDS or symptoms related to HIV infection inde-
pendent of their banding pattern.

We hypothesize that the bands are not diagnostic of AIDS, but seem to correspond
with disease progression, and that they are absent early in the disease, appear later in the
course, and may disappear with advanced disease.

In the healthy state, immunoglobulins are syn-
thesized by many clones of B cells, in response to
the many antigens normally encountered. These
normal immunoglobulins can be separated by
electrophoresis, and will produce a broad, hetero-
geneous band in the immunoglobulin region. In
some disease states, such as hepatobiliary dis-
ease, immunoglobulins originate from the prolif-
eration of one or a few clones of B cells. With
electrophoresis, they produce small, discrete
bands in the immunoglobulin region called oligo-
clonal bands. In some disease states, such as mul-
tiple myeloma, immunoglobulins originate from
the malignant proliferation of a single clone of B
cells and produce large dark bands in the immu-
noglobulin region called monoclonal bands.

Oligoclonal bands have been described in
various conditions, including reticular and non-

From the Division of Hematology/Polly Annenberg Levee He-
matology Center, Department of Medicine, Mount Sinai
School of Medicine (CUNY), New York City. Address reprint
requests to Etta B. Frankel, M.D., 1 Knoll Road, Tenafly, NJ
07670.

reticular malignancy, infectious disease, hepato-
biliary disease, autoimmune disease, heart dis-
ease, and AIDS (1-4). In this study we further
investigated the incidence and significance of
such bands in subjects within the AIDS spectrum
and in patients with hemophilia A. We also mea-
sured, in some cases serially, the immunoglobulin
levels of patients with these illnesses.

Materials and Methods

Serum samples from 224 subjects were ana-
lyzed for the presence of oligoclonal bands. The
samples were obtained from volunteer blood-bank
donors, private ambulatory patients, clinic pa-
tients, and hospitalized patients. Some samples
were frozen and preserved specimens from the
Immunology and Immunochemistry Laboratories
of the Mount Sinai Hospital, New York City.

Determination of Immunoglobulin Concen-
tration, Serum concentrations of IgG, IgA, and
IgM were measured for each sample by the Beck-
man automated immunochemistry system (Beck-
man Instruments, Inc., Brea, CA).

232

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

OLIGOCLONAL BANDING— FRANKEL ET AL.

233

Electrophoresis. The serum samples were
electrophoresed using a high-resolution serum
protein electrophoresis, the Beckman Paragon
System. The serum samples were diluted to an
IgG concentration of 200 mg/dL. A template was
placed on an agarose gel, and 4-6 |jlL of each sam-
ple were applied to the gel. The samples were al-
lowed to diffuse for five minutes and then electro-
phoresed in a barbitol buffer at 100 volts for 40
minutes. The gel was washed in water, fixed in
acid-alcohol composed of 20% acetic acid and 30%
methanol, dried, stained with Paragon Violet,
rinsed in 10% acetic acid, and dried again.

Immunofixation. The bands of 15 subjects
were analyzed to determine the immunoglobulin
class and light-chain type. Sera were diluted in
saline according to their immunoglobulin content
to a final concentration of 200 |xg/dL for IgG and
kappa, 300 fxg/dL for lambda, and 100 |xg/dL for
IgA and IgM. Approximately 4 p,L of diluted sera
were applied to the appropriate lanes on the aga-
rose gel, allowed to diffuse for five minutes, and
then electrophoresed as described above. After
electrophoresis, a template was placed on the gel,
and antisera to IgG, IgA, IgM, and the kappa and
lambda chains were applied to the appropriate
lanes. The antisera were allowed to incubate for
15-30 minutes in a humidified chamber. The gels
were washed in 0.15 M saline and blotted, stained
with Paragon Blue, rinsed in 5% acetic acid, and
dried.

Densitometric Scanning. Selected gels were
scanned using a Hoeffer densitometer (Model
GS300, Hoeffer Scientific Instruments, San Fran-
cisco, CA).

Statistical Analysis. The binomial probabil-
ity distribution was used to determine if the inci-
dence of banding in the patient groups was sta-
tistically different from the 5% incidence of
banding reported in large control groups (5). Fish-
er's exact test with a Bonferoni correction for
multiple comparisons was used to determine if
the incidence of banding was statistically differ-
ent between patient groups.

Results

Table 1 summarizes the results obtained for
controls and for subjects within the AIDS spec-
trum. None of the 40 normal volunteers had
bands, an incidence similar to published normal
control values of 5% (5). Sixteen well homosexual
men who were HIV-antibody-negative had a
banding incidence of 6%, which was not signifi-
cantly different from the published controls. Thir-
teen well homosexual men who were antibody
positive had a banding incidence of 38%; 12 pa-
tients with pre-AIDS-related complex (ARC) (6),
here defined as having persistent generalized
lymphadenopathy, leukopenia, or thrombocytope-
nia but not meeting the accepted criteria for ARC,
had a banding incidence of 50%; 19 patients with
ARC had a banding incidence of 32%; 16 AIDS
patients with Kaposi's sarcoma (KS), with or
without opportunistic infection, had a banding in-
cidence of 44%; and 41 AIDS patients with oppor-
tunistic infections (01) had a banding incidence of
only 15%. All of these groups were significantly
different from the controls (P < 0.01). Fig. 1
shows a sample from a patient with ARC, demon-
strating three bands.

TABLE 1

Oligoclonal Banding in Control Subjects and Patients within the AIDS Spectrum

Diagnosis

No. with
bands (%)

Total no.
of bands

Mean IgG
(g/dL)

Risk group

No. with
bands/N

Controls

Volunteer blood donors 40 0 (0)

AIDS-spectrum Subjects

Well homosexuals

HIV-Ab negative 16 1 (6)

HIV-Ab positive 13 5 (38)

PreARC 12 6 (50)

ARC 19 6 (32)

AIDSiKS 16 7 (44)

AIDS:OI 41 6(15)

1
7

10
8
14
12

0.87*

0.37
<0.01
<0.01

<0.01

0.01

1050 ± 210

1020 ± 230
1645 ± 520
2250 ± 859

2250 ± 1308

3080 ± 1355

1850 ± 575

All homosexual
All homosexual
Homosexual
IVDA

Homosexual
IVDA

Homosexual
Unknown
Homosexual
IVDA-related

1/16

5/13

5/9

1/3

1/13

4/6

3/21

2/12

3/21

1/8

IVDA, intravenous drug abuser; 01, opportunistic infections; KS, Kaposi's sarcoma.
* All groups are compared to a published normal control value of 5% (4).

234

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

B
C

I II

t t t

Fig. 1. Lane A: Normal control. Lane B: Hemophilia
patient treated with large amount of F VIII; 2 bands. Lane
C: ARC, IVDA; 3 bands.

Fig. 3. Development of bands over time in hemophilia pa-
tient treated with large amounts of F VIII. Lanes
A-E: 1978-1982; 0 bands. Lane F: 1984; 2 bands.

Fig. 4. Development of bands over time in hemophilia pa-
tient treated with large amounts of F VIII. Lane A: 03/83;
0 bands. Lane B: 08/83; 0 bands. Lane C: 10/84; 3 faint
bands. Land D: 05/85; 3 distinct bands.

The incidence of banding between patient
groups did not reach statistical significance using
Fisher's exact test with a Bonferoni correction,
which would require a P of 0.01 to be significant.
However, a trend toward significance was sug-
gested between AIDS with 01 and AIDS with KS
(P = 0.03) and between AIDS with 01 and
preARC (P = 0.02).

Mean IgG levels (Table 1) were: controls,
1050 mg/dL; well homosexual antibody-negative,
1020 mg/dL; well homosexual antibody-positive,
1645 mg/dL; preARC, 2250 mg/dL; AIDS with 01,
1850 mg/dL; and AIDS with KS, 3000 mg/dL.

Serial IgG levels were available for 13 of 41
patients who had AIDS with 01 (Fig. 2), and
these were analyzed for changes in IgG level. A
change in IgG level is defined here as greater
than 100 mg/dL. Seven patients had decreasing
IgG levels, one had increasing, then decreasing
IgG levels, 2 had stable levels, and 3 had increas-
ing levels.

In the hemophilia group (Table 2), no band-
ing was seen in nine patients who had never been
treated with factor VIII concentrate (F VIII),
though they may have been treated with cryopre-
cipitate. No banding was seen in five patients
who had been treated with fewer than 10 infu-
sions of F VIII. In patients treated with large
amounts of F VIII (10 infusions or more), the in-
cidence of banding was 50%, which was statisti-
cally different from controls (P < 0.01). In pa-
tients with hemophilia who developed AIDS, the
incidence of banding was 23%, not statistically
different from controls. Fig. 1 shows a sample
from a patient with hemophilia treated with large
amounts of F VIII, demonstrating two bands.

In several patients treated with large
amounts of F VIII, the development of multiple
bands occurred over time, presumably related to
increased amounts of F VIII infused. Serial sam-
ples of patients treated with large amounts of F
VIII (Figs. 3, 4) show the development of in-
creased numbers and intensity of bands over
time.

Mean IgG levels in study subjects tended to
increase with increased use of F VIII:

• no F VIII or cryoprecipitate only: 1190 mg/dL

• treated with F VIII <10 times: 2020 mg/dL

• treated with F VIII ^10 times: 2070 mg/dL

• treated with F VIII ^10 times and developed
AIDS: 2510 mg/dL

Serial IgG levels were available in 3 of the 4 he-
mophiliac patients who developed AIDS (Fig. 5).
All three patients had initially rising and subse-
quently falling IgG levels. There was no change

Vol. 60 No. 3

OLIGOCLONAL BANDING— FRANKEL ET AL.

235

6800
6700
6600

IgG: goo
(mg/di) 700
600

Date 0 1980 81 82 83 84 85 86

Fig. 2. Changes in IgG over time in 13 patients who had
AIDS with opportunistic infections.

in the banding pattern of one hemophiliac AIDS
patient as the IgG level decreased.

Immunofixation was performed on the sera of
eight patients in the AIDS group who had a total
of 13 bands. Six bands were IgG kappa and IgG
lambda and seven were IgG kappa only.

Immunofixation was performed on the sera of
seven patients in the hemophiliac group who had
a total of 12 bands. Five bands were IgG kappa
and IgG lambda, and seven were IgG kappa only.

Discussion

The absence of banding in our controls and
the 6% incidence in well homosexual men who are

HIV-antibody-negative are similar to the inci-
dence reported in normal controls (1). The band-
ing incidence is 30%-50% in well antibody-posi-
tive male homosexuals, and patients who have
preARC, ARC, and AIDS with KS, whereas it is
15% in AIDS with 01. The 15% incidence in pa-
tients who have AIDS with 01 is similar to that
reported by Papadopoulos et al.; the 44% inci-
dence in patients who have AIDS with KS is
somewhat lower than in that series (2). The inci-
dence of banding was significantly less in the
AIDS with 01 group than in well, antibody-posi-
tive homosexual men and patients who have pre-
ARC, ARC, and AIDS with KS. The IgG data of
patients with AIDS with 01 show that a decreas-
ing IgG level is often present; 8 of 13 patients
studied serially had decreasing levels of this im-
munoglobulin. Both the decreased frequency of
banding and the decreasing IgG levels suggest a
progressive B-cell dysfunction as one progresses
within the AIDS spectrum to opportunistic infec-
tion.

In patients with hemophilia A treated with
large amounts of F VIII, the incidence of banding
is 50%. The development of multiple bands is
shown to occur over time in several patients, pre-
sumably related to increased amounts of F VIII
used for treatment. The presence of banding, how-
ever, does not seem to correlate with clinical sta-
tus. Among patients treated with large amounts
of F VIII, 4 of 24 with banding have signs or
symptoms which could fall within the AIDS spec-
trum, including persistent generalized lymphade-
nopathy, fever, weight loss, or idiopathic throm-
bocytopenic purpura. Five of 24 patients without
banding have similar signs of symptoms. The
other 39 patients treated with large amounts of F
VIII are clinically well, except for hemophilia.

TABLE 2

Oligoclonal Banding in Hemophilia A

Subgroup

No. with bands (%)

Total no. of bands

Mean IgG (mg/dL)

Controls
n = 40

Volunteer blood donors 40

Hemophiliac Subjects
n = 67

Cryo only 9
F VIII

(<10 infusions) 5
F VIII

(10 infusions) 49
Developed AIDS 4

0(0)

0 (0)
0(0)

24 (50)

1 (25)

0
0

0.87*

0.63

0.77

<0.01
0.17

1050 ± 210

1190 ± 360

1930 ± 410

2105 ± 805
2510 ± 950

Cryo, cryoprecipitate. F VIII, factor VIII.

* All subgroups are compared to a published normal control value of 5% (4).

236

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

IgG 14

(mg/dl) 13

Date 1 977 78 79 80 81 82 83 84

Fig. 5. Changes in IgG over time in 3 of 4 hemophilia pa-
tients who developed AIDS.

Only one of 4 patients with hemophilia who de-
veloped AIDS had banding. Again, there was a
suggestion that decreasing IgG levels might be
associated with deteriorating clinical status. In 3
patients who developed AIDS and in whom serial
samples were available, the IgG levels initially
increased, and then later decreased.

It has been previously shown by Lane et al (5)
that patients with AIDS have abnormal B-cell
function. They have an increased number of B
cells which spontaneously secrete immunoglobu-
lin, and this finding is more pronounced in pa-
tients with KS than in those with 01. AIDS pa-
tients also have an abnormally reduced B-cell
proliferative response to mitogens and to new an-
tigens. In another study, Lane et al. showed that
patients with 01 are more immunosuppressed
than patients with KS (7), as judged by lower
numbers of total lymphocytes, T lymphocytes,
and B lymphocytes, and decreased proliferative
responses to mitogen (4).

Amadori and Chieco-Bianchi have also sum-
marized B-cell abnormalities in HIV infection (8-
14). B cells are activated in large numbers and
secrete large amounts of immunoglobulin. Some
of the immunoglobulin is polyclonal and directed
against a variety of antigens, but memory B lym-
phocytes do not appear to function normally.
Much of the secreted immunoglobulin, however,
is directed against the HIV virus itself, and is felt
to be the origin of the oligoclonal and monoclonal
bands seen in the serum and cerebrospinal fluid
of these patients. Amadori hypothesizes that
these antibodies may be involved in the patho-
genesis of immune damage seen in HIV-infected
patients by enhancing infectivity, by promoting
complement and cell-mediated damage to in-

fected and bystander cells, and by promoting lym-
phoma development.

The observations we have made are compat-
ible with the findings of Lane and Amadori and
are compatible with, but do not prove, the follow-
ing sequence of events. HIV-infected individuals,
including patients within the AIDS spectrum and
patients with hemophilia treated with frequent F
VIII infusions (presumably 95% of whom are
HIV-antibody positive) are exposed to a variety of
antigens early in the disease. These patients are
initially able to respond appropriately and pro-
duce immunoglobulins with the aid of T lympho-
cytes. As the disease progresses somewhat, there
is ongoing antigenic stimulation with both regu-
lated and spontaneous secretion of immunoglob-
ulins, leading to hypergammaglobulinemia, in-
creased anti-HIV antibodies, and increased
likelihood of oligoclonal banding. As the disease
progresses even further, the patients develop
more advanced immunodeficiency, with decreas-
ing immunoglobulin levels and decreased oligo-
clonal banding. This last stage may be due to a
profound loss of helper-inducer T lymphocytes,
necessary for B-cell function, or to an intrinsic
dysfunction of the B lymphocyte itself. Further-
more, 01 may ensue most easily in those who
have progressed to this most advanced stage of
B-cell and T-cell immunodeficiency.

Thus, the evaluation of oligoclonal banding
in AIDS and hemophilia may provide another in-
sight into the immune dysfunction of HIV dis-
ease. The bands are not diagnostic of AIDS, as
they are present in several other conditions, but
seem to correspond with the status of disease pro-
gression. They are absent early in the disease,
appear later in the course, and may disappear as
the disease becomes quite advanced.

Acknowledgments

The authors wish to acknowledge the help of Melissa Brener;
Gail Macavy, statistics.

References

1. Kelly RH, Hardy TJ, Shah PM. Benign monoclonal gam-

mopathy: a reassessment of the problem. Immunol In-
vest 1985; 14(3):183-197.

2. Keskhegian AL. Prevalence of small monoclonal proteins

in the serum of hospitalized patients. Am J Clin Pathol
1982; 77(4):436-^42.

3. Papadopoulos NN, Lane HC, Costello R, et al. Acquired

immunodeficiency syndrome. In: Clinical immunology
and immunodeficiency syndrome, 1985, 43^6.

4. Papadopoulos NM, Ellin RJ, Wilson DM. Incidence of oli-

goclonal banding in a healthy population by high-reso-
lution electrophoresis. Clin Chem 1982; 28(4):707-708.

5. Lane CH, Masur H, Edgar LC, Whalen G, Rook AH, Fauci

AS. Abnormalities of B-cell activation and immunoreg-

Vol. 60 No. 3

OLIGOCLONAL BANDING— FRANKEL ET AL.

237

ulation in patients with the acquired immunodeficiency
syndrome. N Engl J Med 1983; 309:453.

6. DeVita VT, Jr, Hellman S, Rosenberg SA. AIDS: etiology,

diagnosis, treatment, and prevention. New York: J.B.
Lippincott, 1985:3.

7. Lane HC, Masur H, Gelmann EP, et al. Correlation be-

tween immunologic function and clinical subpopula-
tions of patients with the acquired immunodeficiency
syndrome. Am J Med 1985; 78:417.

8. Amadori A, Chieco-Bianchi L. B-cell activation and HIV-1

infection: Deeds and misdeeds. Immunol Today 1990;
ll(10):374-9.

9. Amadori A, et al. HIV-1-specific B-cell activation. J Im-

munol 1989; 143(7):2146-2152.
10. Amadori A, et al. In-vitro production of HlV-specific an-
tibody in children at risk of AIDS. Lancet 1988; (Apr
16):852-854.

11. Gallo P, et al. Central nervous system involvement in

HIV infection. AIDS Research And Human Retrovirus-
es 1988; 4(3):21 1-221.

12. Ng L, et al. The clinical significance of human immuno-

deficiency virus type 1 -associated paraproteins. Blood
1989; 74(7):2471-5.

13. Ng VL, et al. High titer anti-HIV antibody reactivity as-

sociated with a paraprotein spike in a homosexual male
with AIDS related complex. Blood 1988; 71(5):1397-
1401.

14. Pahwa, Savita et al. In vitro synthesis of human immu-

nodeficiency virus-specific antibodies in peripheral
blood lymphocytes of infants. Proc Natl Acad Sci USA
1989; 86:7532-7536.

Submitted for publication June 1991.

Revision received May 1992.

Gold-Induced Colitis

A Case Report and Review of the Literature

Victoria Teodorescu, M.D., Joel Bauer, M.D., Simon Lichtiger, M.D., and Mark Chapman, M.D.

Abstract

A case of severe colitis requiring subtotal colectomy following administration of 35 mg
Solganal b for intractable arthritis is described. Abdominal pain and watery diarrhea
developed six weeks after the last dose of gold. Colonoscopy revealed mucosal edema and
ulceration of the entire colon. Supportive measures failed and the patient required sub-
total colectomy. Review of the literature revealed 29 cases, ranging in severity from
limited ileal involvement to fulminant panenteritis. Most of the patients responded to
intravenous fluids, steroids, and antibiotics, but four required surgery. The case described
is notable for the delay in appearance of abdominal symptoms following the cessation of
gold therapy. The mechanism of injury is unknown. Abdominal complaints in a patient
who has received gold therapy, especially parenteral, merit strict attention, even if occur-
ring several weeks after the final dose, and the diagnosis of gold colitis should be enter-
tained.

Most physicians using gold therapy to treat pa-
tients with rheumatoid arthritis are aware of the
significant incidence of toxicity, ranging from
25% to 50% (1). Cutaneous reactions are the most
common, but blood dyscrasias and proximal tubu-
lar damage may also occur, requiring regular ex-
amination of the skin, blood, and urine during
treatment. Gastrointestinal complications other
than nausea and vomiting are extremely rare,
however; only 29 cases of gold enteritis have been
previously reported.

We describe a patient who developed severe
colitis after administration of gold for intractable
rheumatoid arthritis and who ultimately re-
quired subtotal colectomy.

Case Report

A 56-year-old woman was diagnosed with se-
ronegative rheumatoid arthritis six years ago.
Three months prior to admission, she had been

From the Departments of Surgery and Medicine (Gastroenter-
ology), Mount Sinai School of Medicine (CUNY) and The
Mount Sinai Hospital, New York, New York. Address reprint
requests to Joel Bauer, M.D., 25 East 69th Street, New York,
NY 10021.

238

started on weekly gold injections for intractable
arthritis. After two injections, totaling 35 mg, she
developed a rash over her legs. The gold therapy
was discontinued and the rash improved with top-
ical steroids. Six weeks later, however, she devel-
oped fever, abdominal pain, and diarrhea, requir-
ing hospitalization. The pain was distributed
diffusely over the lower abdomen and relieved by
defecation. The diarrhea occurred 10-15 times a
day, was watery to semisolid, and bloody on some
occasions. On colonoscopy, the mucosa was edem-
atous, granular, and friable throughout the entire
colon; cobblestoning and linear ulcers were also
noted. Biopsy revealed severe acute colitis with
many crypt abscesses and mucin depletion. The
patient was given intravenous steroids for three
weeks with no improvement. She was then trans-
ferred to The Mount Sinai Hospital.

On examination, she appeared ill, and had a
pulse rate of 120/min, blood pressure of 90/60
mmHg, temperature of 100°F, and mild pedal
edema. An erythematous, macular, scaly rash
was distributed mainly over her back and arms.
Thrush was noted within the oral cavity. The ab-
domen was distended and diffusely tender with
diminished bowel sounds.

Laboratory data were as follows: The hemo-

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Fig. 1. Right Abdominal films revealing dilated edema-
tous small bowel in midabdomen and loss of haustral mark-
ings in sigmoid colon. Fig. 2. Below left Subtotal co-
lonic resection specimen showing active universal ulcerative
colitis. Fig. 3. Below right Closer view showing marked
edema, granular mucosa, and linear ulcers.

240

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

globin was 8 g/dL and hematocrit 23.4%; white
blood cell count 8200/mm with 23% neutrophils,
50% band forms, 25% lymphocytes, and 2% mono-
cytes. Erythrocyte sedimentation rate was ele-
vated to 72 mm/hr. Prothrombin time was nor-
mal. Serum sodium was 117 mEq/L, potassium
4.4 mEq/L, chloride 95 mEq/L, CO2 19 mEq/L,
BUN 13 mg/dL. Total protein was decreased to
3.2 g/dL, with serum albumin only 1.2 g/dL. Bil-
irubin was 4.2 mg/dL and amylase 142 I units/
mL; all other liver function tests were normal. An
abdominal X-ray revealed dilated loops of large
and small bowel (Fig. 1).

Her condition continued to deteriorate, and
two days after transfer she underwent subtotal
colectomy and Brooke ileostomy (Figs. 2, 3). At
the time of operation, the entire intraabdominal
colon was inflamed and edematous, but the ter-
minal ileum and rectum appeared normal.

Postoperatively, the patient developed some
melena. Upper gastrointestinal endoscopy
showed two ulcers, one in the pyloric region and
another on the lesser curvature of the stomach.
She was treated with cimetidine. The remainder
of her hospital stay was unremarkable; approxi-
mately 11/2 months after admission, she was dis-
charged. She was readmitted six months later for
elective ileorectal anastomosis and has remained
well over the next 3V2 years of follow-up.

Discussion

In 1935, Goldhammer described a 47-year-old
woman who developed severe diarrhea after re-
ceiving a total dose of 240 mg Solganal b for pri-
mary chronic polyarthritis. The patient died after
12 days despite supportive therapy. Autopsy
showed severe mucosal edema and hemorrhage
throughout the small and large bowel (3). Since
then, 29 cases of gold-induced enterocolitis, in-
cluding this one, have been described.

Most of these patients, like this one, have
been middle-aged women, but this observation
probably reflects the greater prevalence of rheu-
matoid arthritis in this group rather than an in-
herent susceptibility.

Our case was unusual in three respects.
First, six weeks had elapsed after the cessation of
gold therapy before the onset of abdominal com-
plaints. In most cases, enterocolitis occurs early
in the course of treatment. All patients described
had received 510 mg or less, except for one who
had received two cycles totaling 3,200 mg before
enterocolitis developed (15).

Second, at the time of transfer, the patient
still had the skin rash that had been her initial

symptom. Only four reported cases had other
signs or symptoms of gold toxicity accompanying
their enterocolitis (3, 8, 11, 12).

Third, the need for colectomy was unusual.
Although 6 of the 12 patients described up to 1976
died (3-12), only 2 of the 17 patients subsequently
reported on have died (13-27). Only three pa-
tients required colectomy, two for perforation (15,
18). One survivor underwent diversion with a
loop ileostomy (26).

The mechanism of gold-induced bowel injury
is currently unknown. Although nearly all pa-
tients received parenteral therapy, one case did
occur in a patient taking auranofin, an oral com-
pound. Thus, a direct toxic effect on the intestinal
mucosa related to the drug's high rate of fecal
excretion may be the inciting event (2). Two pa-
tients, however, both of whom died, had a history
of bloody diarrhea (7, 9); in these patients, gold
therapy may have exacerbated previously exist-
ing disease. In all reported cases, the joint disease
was characteristic of rheumatoid arthritis and
not suggestive of colitic arthritis. In all previ-
ously reported cases, stool examination was per-
formed and no pathogens were identified. Small-
bowel series have revealed rapid transit time (9,
13, 20) or mucosal thickening (8, 11). Barium en-
emas have shown colonic spasticity (3), edema-
tous mucosa (9, 14, 22), or a normal appearance
(10, 11, 19, 20). Involvement ranged from ileum
only (7) to fulminant panenteritis (5, 8, 18).
Pathologic changes included edematous mucosa,
ulcerations, and hemorrhages with infiltration of
lymphocytes, plasma cells, and polymorphonu-
clear lymphocytes (3, 5, 9-19, 21-23). One of the
patients described by Fam et al. also had throm-
bosis of several mucosal vessels, identified on his-
tologic review (15).

Alternatively, the mechanism of injury may
be an immune-mediated response. In support of
this theory, Szpak et al. demonstrated marked
tritiated thymidine uptake following mixed lym-
phocyte stimulation with gold sodium thiomalate
in the patient they described, a phenomenon not
seen in normal controls or arthritic patients re-
ceiving gold therapy without evidence of toxicity
(13). In addition, Wright et al. demonstrated cir-
culating immune complexes in their patient,
which were cleared following the resolution of
colitis (25). A delayed immune response rather
than direct mucosal toxicity may account for the
six- week lag in onset of abdominal symptoms and
the persistent rash in our patient.

In any case, early diagnosis is mandatory.
The onset of diarrhea in a patient receiving or
who recently received parenteral gold should

Vol. 60 No. 3

GOLD-INDUCED COLITIS— TEODORESCU ET AL.

241

alert the physician to the possibility of this com-
plication. Aggressive investigation to determine
the cause should be instituted and gold therapy
discontinued immediately, if it has not been al-
ready. Most patients will improve with adequate
fluid replacement and administration of steroids
and antibiotics. If this regimen fails, colectomy
may be required. In contrast, patients receiving
gold orally will frequently experience mild diar-
rhea or other gastrointestinal symptoms. Such re-
sponses in these patients require downward ad-
justment of the dose and close follow-up.

References

1. Oilman AG, Goodman LS, Oilman A. Goodman and Gil-

man's the pharmacological basis of therapeutics. New
York: Macmillan, 1980:713-717.

2. Gottlieb NL. Gold excretion and retention during aurano-

fin treatment: a preliminary report. J Rheumatol 1979;
6(Suppl 5):61-67.

3. Goldhammer S. A fatal case of solganol poisoning. Med

Klin 1935; 31:645-647.

4. Perry MW. Gold injections and colitis. JAMA 1939; 113:

965 (letter).

5. Anderson NL, Palmer WL. The danger of gold salt ther-

apy, report of a fatal case. JAMA 1940; 115:1627-1630.

6. Kandrac M, Pav J, Pechova I. Successful steroid therapy

of severe enteritis caused by gold therapy. Cas Lek
Cesk 1961; 100:361-367.

7. Kaplinsky N, Pras M. Ulcerative colitis and gold treat-

ment in rheumatoid arthritis. Harefuah 1971; 80:406-
407.

8. Roe M, Sears AD, Arndt JH. Gold reaction panenteritis,

case report with radiographic findings. Radiology 1972;
104:59-60.

9. Kaplinsky N, Pras M, Frankl O. Severe enterocolitis com-

plicating chrysotherapy. Ann Rheum Dis 1973; 32:574—
577.

10. Stein HB, Urowitz MB. Gold-induced enterocolitis. Case

report and literature review. J Rheumatol 1972; 3:21-
26.

11. Siegmen-Ingra Y, Yaron M, Siletzki M, et al. Colitis and

death following gold therapy. Rheum Rehabil 1976; 15:
245-247.

12. Gerster JC, deKalbermatten A, et al. Reactions toxiques

aux sels d'or avec enterocolite grave chez un homme
atteint d'une polyarthrite rhumatoide. Schweitz Med
Wochenschr 1976; 106:1606-1608.

13. Szpak MW, Johnson RC, Brady CE, et al. Gold (Au) in-

duced enterocolitis. Gastroenterology 1979; 76:1257
(abstr).

14. Sckolnick BR, Katz LA, Kozower M. Life-threatening en-

terocolitis after gold salt therapy. J Clin Gastroenterol
1979; 1:145-148.

15. Fam AG, Paton TW, Shamess CJ, Lewis AJ. Fulminant

colitis complicating gold therapy. J Rheum 1980; 7(4):
479^85.

16. Huston GJ. Gold colitis, therapy and confirmation of mu-

cosal recovery by measurement of rectal potential dif-
ference. Postgrad Med J 1980; 56:875-876.

17. Martin DM, Goldman JA, Gilliam J, Nasrallah SM. Gold-

induced eosinophilic enterocolitis: response to oral cro-
molyn sodium. Gastroenterology 1981; 80(6):1567-
1570.

18. Eaves R, Hansky J, Wallis P. Gold induced enterocolitis:

case report and a review of the literature. Aust NZ J
Med 1982; 12:617-620.

19. Susenik S, Hirsch M, Yanai-Inbar I, Krawiec J, Freund B,

Horowitz Y. Enterocolitis complicating chrysotherapy:
case report and review of the literature. Isr J Med Sci
1982; 18:1040-1043.

20. Nagler J, Paget SA. Nonexudative diarrhea after gold salt

therapy: case report and review of the literature. Am J
Gastroenterol 1983; 78:12-14.

21. Jarner D, Nielsen AM. Auranofin (SK + F 39612) in-

duced enterocolitis in rheumatoid arthritis: case report.
Scand J Rheumatol 1983; 12:254-256.

22. Manigand G, Dumont D, Faux N, et al. Colite aigue au

cours des traitements pars sels d'or. Presse Med 1983;
12:2112-2113.

23. White RF, Major GAC. Gold colitis. Med J Aust 1983;

1:174-175.

24. Reinhart WH, Kappeler M, Haiter F. Severe pseudomem-

branous and ulcerative colitis during gold therapy. En-
doscopy 1983; 15:70-72.

25. Wright A, Benfield GFA, Felix-Davies D. Ischemic colitis

and immune complexes during gold therapy for rheu-
matoid arthritis. Ann Rheum Dis 1984; 43:495^97.

26. Jackson CW, Haboubi NY, Whorwell PJ, Schofield PF.

Gold-induced enterocolitis: case report and review. Gut
1986; 27:452-456.

27. Lee FY, Lin HY, Pan S. Gold-induced fulminant colitis in

a patient with psoriatic arthritis. J Clin Gastroenterol
1988; 10(1):116-117.

Submitted for publication March 1991.

Revision received March 1992.

Progressive Nemaline Rod Myopathy in a
Woman Coinfected with HIV-1
and HTLV-2

Joseph Maytal, M.D., Steven Horowitz, M.D., Stanley Lipper, M.D., Bernard Poiesz, M.D.,
Chang Yi Wang, Ph.D., and Frederick P. Siegal, M.D.

Abstract

Nemaline-rod myopathy was recently reported in eight young males infected with human
immune deficiency virus type 1 (HIV-1). A 41-year-old woman had a 2-year history of
progressive proximal-muscle weakness. Muscle biopsy demonstrated the presence of nem-
aline rods, predominantly in type 1 fibers. She was coinfected with HIV-1 and HTLV-2, as
evidenced by positive polymerase chain reaction and serology. There was no lymphopenia
or CD4 lymphopenia, despite an abnormal T-cell subset ratio, high CDS count, skin an-
ergy, and depressed in vitro response to mitogens. This case raises the possibility that dual
infection may play a role in the pathogenesis of the rare nemaline-rod myopathies of
HIV-infected patients.

Skeletal-muscle disease associated with human
immunodeficiency virus type 1 (HIV-1) infection
has recently been reported. Most reported cases
confirmed by muscle biopsy exhibit inflammation
as seen in polymyositis; others show features of
necrosis and regeneration without concurrent cel-
lular infiltrates, but have nonetheless been con-
sidered inflammatory in origin (1-4). In addition,
over the last four years, eight cases, all young
males infected with HIV-1, have been reported
with progressive proximal weakness; in all the
predominant finding on muscle biopsy was the
presence of nemaline rods (4—8). We report the
case of a young woman with severe progressive
proximal weakness, coinfected with HIV-1 and
human T-cell lymphotropic virus 2 (HTLV-2), in

From the Departments of Neurology (SH), Pathology (SL),
Medicine (FPS), and the Division of Pediatric Neurology,
Schneider Children's Hospital (JM), Long Island Jewish Med-
ical Center, the Long Island Campus of Albert Einstein Col-
lege of Medicine; Department of Medicine, SUNY Health Sci-
ences Center (BP), Syracuse, NY, and United Biomedical, Inc.
(CYW), Lake Success, NY. Address reprint requests to Joseph
Maytal, M.D., at the Division of Pediatric Neurology, Schnei-
der Children's Hospital, Long Island Jewish Medical Center,
New Hyde Park, NY 11042.

whom a nemaline-rod myopathy was seen on bi-
opsy.

Case Report

In June 1990 a 41-year-old U.S. -born black
woman consulted a physician for a 2-year history
of progressive weakness which began with diffi-
culty climbing stairs and getting off a toilet seat
and was followed by proximal upper-extremity
weakness six months later. She reported some in-
termittent, pruritic nodular lesions on her legs.
She had had dermatomal herpes zoster in October
1989. One relevant element in her history was
that she had been a high-school track star. Both
her parents, eight siblings, and four children
have no evidence of neuromuscular problems. Her
husband, an intravenous-drug user, had died of
AIDS in September 1988. She had not received
blood transfusions or used illicit drugs.

She had a markedly waddling gait with
prominent lumbar lordosis and was unable to rise
from a chair. There was bilateral and symmetri-
cal weakness of the neck flexors (4/5); deltoid, su-
praspinatus and infraspinatus (2/5); biceps, tri-
ceps, and wrist extensors (3/5); finger flexors and
intrinsic hand muscles (4/5); iliopsoas (1/5), quad-

242

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

MYOPATHY WITH HIV-1 AND HTLV-2— MAYTAL

243

riceps, hamstrings, and glutei (2/5); and all mus-
cles distal to the knees (5-/5). All deep tendon
reflexes were absent except the ankle jerks,
which were 2 + . The general physical examina-
tion was negative; in particular, lymphadenopa-
thy, splenomegaly, and oral abnormalities were
absent.

Creatinine kinase (CK) was normal (<190
lU/L) on five separate occasions. The ESR was 98
mm/hr. Antibodies tested for HIV-1 by enzyme-
linked immunoadsorbent assay (ELISA) and
Western blotting were positive (Table 1). Peptide
ELISA for HTLV-1 and HTLV-2 were positive,
but the reactivity was far stronger to HTLV-2
peptides than to those derived from HTLV-1, with
which there is a known cross-reactivity. The pres-
ence of both HTLV and HIV-1 proviruses was con-
firmed by the polymerase chain reaction (PGR),
using probes for HIV-1 gag (6500 copies/10® cells),
HTLV-1/2 pol SKI 15, and HTLV-2 pol SK188;
the HTLV-l-specific pol SKI 12 probe did not re-
act on PGR.

These results, combined with the serologic
findings on peptide ELISA, suggest that the pa-
tient is infected with HTLV-2 and HIV-1. Direct
cocultures of the patient's cells with PHA-stimu-
lated normal cells for HIV-1, using HIV-1 p24 an-
tigen as the culture endpoint, were negative
through 27 days. The patient was anergic to a
battery of seven common recall antigens in de-
layed-type hypersensitivity (Multitest-GMI,
Merieux Institute USA, Miami, FL).

Electromyographic analysis of the motor-
unit potentials of one vastus medialis, anterior
tibialis, and the biceps brachii muscle revealed
decreased mean durations, reduced amplitudes,
and minimal polyphasia. Muscle biopsy was ob-
tained from the opposite vastus medialis. Hema-
toxylin- and eosin-stained sections showed a
highly abnormal muscle with marked variation
in fiber diameter (range 5-140 micron). Large fi-
bers were oval to rounded. Internal nuclei were
markedly increased in size. Atrophic rounded or
wedged fibers were noted singly and in small
groups. Increased fibrous tissue was present
around atrophic fibers. Atrophic fibers with nu-
clear clumps were present. Scattered fibers
showed vacuolation and granular degeneration.
A few small endomysial, perimysial, and epimy-
sial vessels showed light lymphoplasmacytic cuff-
ing. Muscle trichrome stain revealed numerous
aggregates of nemaline rods mainly within atro-
phic fibers. Enzyme histochemistry (ATPase) re-
vealed that nemaline rods were present mainly in
type 1 fibers, although type 2 fibers were also
involved to a lesser degree. Electromicroscopy

showed the presence of nemaline rods in varying
profusion, mostly in atrophic fibers, together with
Z-band streaming, myofilament loss, and disorga-
nization (Fig.).

Discussion

Over the last four years, eight cases of nem-
aline-rod myopathy have been reported in associ-
ation with HIV-1 infection, all in young males
with a history of progressive proximal-muscle
weakness (4-8) (Table 2). In our patient a nema-
line-rod myopathy occurred in association with
coinfection by HIV-1 and HTLV-2, thus raising
the possibility that dual infection may play a role
in its pathogenesis.

In all reported cases, including ours, the elec-
trophysiologic studies were consistent with a my-
opathy. Creatinine kinase levels were mildly el-
evated in 5 patients, within normal limits in 3,
and not available in one (Table 2). Helper-sup-
pressor ratio was abnormally low in 3 patients
(6), as in the case reported here, normal in 1 (6),
and not reported in 4 patients. CD4 counts were
reported in 3 patients and varied from 951 mm^,
in our patient, to 540 mm^ and 560 mm^. HTLV-
infection status was not evaluated in the other
cases. Of the 8 patients with nemaline-rod myop-
athy so far reported, 5 had not developed any
HIV-related symptoms (4-6). Two others had
AIDS-related complex (ARC) (4, 8), and only one
exhibited fully developed AIDS (7) (Table 2).

The most striking histologic finding on mus-
cle is the diffuse presence of nemaline rods; in five
cases, including ours, these rods were predomi-

TABLE 1

Summary of Patient's Immune Function Tests

Test

Results

Absolute lymphocyte count

2,438 cells/mm^

CD4 count

951 cells/mm^ (39%)

CDS count

l,329/mm3 (54%)

CD4:CD8 ratio

0.72

Lymphocyte proliferation

without added mitogen

1809 cpm (elevated)

Response to phytohemagglutinin

18,037 cpm (depressed)

Response to pokeweed mitogen

1,872 cpm (depressed)

Response to concanavalis A

12,861 cpm (normal)

Natural killer cell activity

against K562 target cells

depressed

Antibodies to rapid plasma

reagin (RPR)

not detectable

ELISA for cytomegalovirus

and toxoplasma

positive

Rheumatoid factors

(serum dilution of 1/20)

present

Antinuclear antibodies (ANA)

negative

244

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

TABLE 2

Findings in 9 Patients with HIV -Associated Nemaline Myopathy

Muscle biopsy

Patient Risk Associated Clinical CPK Treatment &

age/sex/ref factor condition signs (lU/L) Inflammation Nemaline rods course

26/My5 H

HIV^

PMW

480 ( t )

26/M/6 H

40/M/4

34/M/7

HIV

29/M/6 Bisexual HIV^

32/M/4 H

45/M/4 H

ARC

HIV^

HIV"

Unknown AIDS

24/M/8 IVDA ARC

41/F/ Husband HIV*

present case IVDA

Died from HTLV-2 "
AIDS

PMW upper

> lower

PMW upper

> lower

PMW
decreased
reflexes

PMW

752 ( t )
900 ( t )
543 ( t )

260 (NL)

Severe

PMW

Not Avail.

PMW, 210 (NL)

dysphagia

PMW

313 ( t )

142 (NL)

Predominantly
type I fibers

Not treated
Slow

progression

of weakness

and wasting

(14 mo)
Pred; clinical

improvement

(6 mo)
Plasmaphoresis;

minimal

improvement
Pred; improved

for 2 yr.;

relapsed;

pred.

increased
Pred for 2 yr;

improved;

normal and

off

medication
(5 yr)
Pred; improved
(3 yr)

spontaneous
improvement
(4 mo)

Retrovir for
AIDS; when
PMW
developed,
interferon
alpha was
added; some
improvement
(2 yr)

Not treated

Progressive
deterioration
(6 mo)

Not treated

Stable (20 mo)

CPK, creatine phosphokinase; H, homosexual; AIDS, acquired immunodeficiency syndrome; ARC, AIDS-related complex; PMW,
proximal muscle weakness; Pred, prednisone; IVDA, intravenous drug abuse; + , positive.

nantly present in the type 1 fibers (5, 6, 8). Ne-
crosis and loss of sarcomeric thick filaments have
also been reported in most cases. Inflammatory
infiltrates were conspicuously absent in 6 pa-
tients (including the present case), mild in 2, and
moderate to severe in one (4).

Nemaline rods originate in the Z bands (9),
and their constituents are actin and alpha-actin,
the protein through which actin filaments are at-
tached to the Z bands. They are the result of a

nonspecific myofibrillar alteration resulting from
Z-band disorganization and disruption. Rod bod-
ies were first described in congenital nemaline
(rod body) myopathy (10). Adult-onset nemaline
myopathy (11) has also been described. Nemaline
rods have also been reported in other disorders,
including mitochondrial myopathies (12) and poly-
myositis (13).

The pathogenesis of nemaline rods in HIV-
related myopathy is not known; the finding of in-

Vol. 60 No. 3

MYOPATHY WITH HIV-1 AND HTLV-2— MAYTAL

245

flammatory changes, severe in one patient and
mild in 2 (4), may link these patients with those
in whom the myopathy is clearly inflammatory
and in whom nemaline rods were not found. Simp-
son and Bender (4) comment that a low degree of
inflammatory response may be related to T- and
B-cell abnormalities. It is possible that all the
myopathic variations in HIV infection reflect a
spectrum of the same disease, predominance of
nemaline rods being at one extreme and predom-
inance of inflammatory infiltrates at the other
(4, 8).

The finding of progressive nemaline (rod)
myopathy as the only clinical manifestation re-
ported in some HIV-related myopathies — with
complete failure to demonstrate or cultivate the
virus from muscle specimens using a variety of
techniques, including electronmicroscopy — sug-
gests that retrovirus can indirectly, by means of
immunopathologic mechanisms, affect the pri-
mary structural and contractile proteins of the
muscles (5). It is possible that this HIV-associated
myopathy is produced by an autoimmune mecha-
nism; autoimmune thrombocytopenic purpura
and chronic inflammatory neuropathy are exam-
ples of disorders considered to be autoimmune
that occur in association with HIV infection (14,

15). In support of this hypothesis is the improve-
ment of the symptoms with corticosteroid in 4 pa-
tients (4, 6) and plasmapheresis in 1 (6) (Table 2).
On the other hand, Wiley et al. (16) postulated
direct infection of muscle fibers by HTLV-1 in a
patient who had polymyositis and was doubly in-
fected with HTLV-1 and HIV. Ishii et al. (17) re-
ported the isolation of HTLV-1 from a patient
with polymyositis. Further studies are required
to elucidate the pathogenesis of these retrovirus-
associated myopathies.

In the patient reported on here, the clinical
significance of the coinfection with HIV-1 and
HTLV-2 in causing the myopathy is not clear.
The spectrum of HTLV-l-associated neurologic
disease is restricted mostly to the tropical spastic
paraparesis syndrome (18) and to a few cases of
polymyositis (16, 17). No neurologic syndrome re-
lated to HTLV-2 has yet been described, although
a few sporadic cases of central nervous system
disorders, for example cerebral lymphoma, sub-
acute encephalitis (19), uveomeningoencephalitis

(20) , and dementia with frontal lobe syndrome

(21) , were reported in HTLV-2-positive patients.
Recently, a case of a patient infected with
HTLV-2 has been associated with a tropical spas-
tic paraparesis-like illness (22).

Fig. Electron micrograph of nemaline rods in 41-year-old woman coinfected with HIV-1 and HTLV-2 (original magni-
fication X 17,500).

246

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

In the case reported here, the relevance of the
disturbances in cellular immunity (CDS lympho-
cytosis with abnormal T-cell subset ratio) to her
initial clinical symptoms and myopathic findings
is also uncertain. It is possible that high CDS
counts are involved in the pathogenesis of this
patient's myopathy. The possible role of poly-
clonal or perhaps even monoclonal myocytotoxic
T cells in the myopathic process clearly deserves
further evaluation.

This case suggests that serologic surveys of
primary myopathies in HIV infection for the pres-
ence of the other known human retroviruses
(HTLV-1 or -2) should be performed; it is possible
that dual infection is a necessary pathogenetic
factor.

References

1. Dalakas MC, Pezeshkpour GH, Gravell M, Sever JL. Poly-

myositis associated with AIDS retrovirus. JAMA 1986;
256:2381-2383.

2. Stern R, Gold J, DiCarlo EF. Myopathy complicating the

acquired immune deficiency syndrome. Muscle Nerve
1987; 10:318-322.

3. Lange DJ, Britton CB, Younger DS, Hays AP. The neu-

romuscular manifestation of HIV. Arch Neurol 1988;
45:1084-1088.

4. Simpson DM, Bender AN. Human immunodeficiency vi-

rus-associated myopathy: analysis of 11 patients. Ann
Neurol 1988; 24:79-84.

5. Dalakas MC, Pezeshkpour GH, Flaherty M. Progressive

nemaline (rod) myopathy associated with HIV infec-
tion. N Engl J Med 1987; 317:1602-1603.

6. Gonzales MF, Olney RK, So YT, et al. Subacute structural

myopathy associated with human immunodeficiency vi-
rus infection. Arch Neurol 1988; 45:585-587.

7. Geny C, Poli F, Mhiri C, Rekacewicz I, Gherardi R, Revuz

J. Myopathie a batonnets chez un sujet atteint du SIDA.
Presse Med 1988; 18:2021.

8. Cabello A, Martinez-Martin P, Gutierrez-Rivas E, Madero

S. Myopathy with nemaline structures associated with
HIV infection. J Neurol 1990; 237:64-66.

9. Yamaguchi M, Robson RM, Stromer MH, et al. Actin fil-

aments from the backbone of nemaline myopathy rods.
Nature 1978; 271:266-267.

10. Shy GM, Engel WK, Samero JE, Wanko T. Nemaline my-

opathy: a new congenital myopathy. Brain 1963; 86:
793-810.

11. Brownwell AKW, Gilbert JJ, Shaw DT, et al. Adult-onset

nemaline myopathy. Neurology 1978; 28:1306-1309.

12. Fukunaga H, Osame M, Igata A. A case of nemaline my-

opathy with ophthalmoplegia and mitochondrial abnor-
malities. J Neurol Sci 1980; 46:169-177.

13. Cape CA, Johnston WW, Pitner SE. Nemaline structure

in polymyositis. Neurology 1970; 20:494-502.

14. Walsh C, Krigel R, Lennette E, Karpatkin S. Thrombocy-

topenia in homosexual patients. Ann Intern Med 1985;
103:542-545.

15. Cornbluth DR, McArthur JC, Kennedy ME, et al. Inflam-

matory demyelinating peripheral neuropathies associ-
ated with human T-cell lymphotropic virus type III in-
fection. Ann Neurol 1987; 21:32-^0.

16. Wiley CA, Nerenberg M, Cros D, et al. HTLV-1 polymy-

ositis in a patient also infected with the human immu-
nodeficiency virus. N Engl J Med 1989; 320:992-995.

17. Ishii K, Yamato K, Iwahara Y, et al. Isolation of HTLV-1

from muscle of a patient with polymyositis. Am J Med
1992; 90:267-269.

18. Vernant JC, Maurs L, Gessin A, et al. Endemic tropical

spastic paraparesis associated with human T lympho-
tropic virus type I: a clinical seroepidemiological study
of 25 cases. Ann Neurol 1987; 21:123-130.

19. Clavel F, Mansinho K, Chamaret S, et al. Human immu-

nodeficiency virus type 2 infection associated with
AIDS in West Africa. N Engl J Med 1987; 19:1180-
1184.

20. Hormigo A, Brave-Marques JM, Souza-Ramalho P, et al.

Uveomeningoencephalitis in a human immunodeficien-
cy virus type 2 seropositive patient. Ann Neurol 1988;
23:308-310.

21. Livrozet JM, Ninet J, Vighetto A, et al. One case of HIV-2

AIDS with neurological manifestations. J Acquir Im-
mune Defic Syndr 1990; 3:927-928.

22. Berger JR, Svenningsson A, Raffantis, et al. Tropical

spastic paraparesis-like illness occurring in a patient
dually infected with HIV-1 and HTLV-2. Neurology
1991; 41:85-87.

Submitted for publication August 1992.
Revision received December 1992.

Abstracts

The abstracts whose titles and authors are presented here — all authors are principal investi-
gators in the Samuel Bronfman Department of Medicine of the Mount Sinai School of Medi-
cine, which includes the affiliates Beth Israel Medical Center, the Bronx Veterans Adminis-
tration Medical Center, and Elmhurst Hospital Center, in New York City— were exhibited with
the abstract texts as posters in the Department of Medicine Tenth Annual Research Day,
November 17, 1992.

Six abstracts by students, fellows, and house staff working in the Department of Medicine
received Tenth Annual Department of Medicine Research Awards and were presented orally at
a Special Grand Rounds the same day; these are identified by asterisks.

Abstracts in Medicine from
Mount Sinai Medical Center 1992

Cardiology

1. Tc99m Teboroxime SPECT Myocardial "Redistribution" Perfusion
Imaging. M. Henzlova, J. Machac

2. Clinical Correlates of Increased Lung Uptake of I-123-Metaiodo-
benzylguanidine in Ischemic Heart Disease. J. Machac, S. Vallabhajo-
sula, J. Gatley, D. Schyer, N. Volkow, A. Wolf, S. Goldsmith, J. A.
Cromes, R. Gorlin

3. Effects of Ischemia at Rest on Myocardial Sympathetic Nerve In-
tegrity in Humans. J. Machac, S. Vallabhajosula, J. Gatley, D. Schlyer,
N. Volkow, A. Wolf, S. Goldsmith, J. A. Gomes, R. Gorlm

4. Polycystic Ovary Syndrome: Lack of Hypertension Despite Pro-
found Insulin Resistance. R. Phillips, S. Zimmermann, A. Dunaif, D.
Finegood, L. Krakoff

5. Angioplasty Activates the Clotting Cascade in Patients with Com-
plex Coronary Lesions. N. Khaghan, J. Marmur, K. Bauer, S. Sharma,
D. Israel, J. Ambrose

Clinical Immunology

6. Biological Activities of Polyethylene-Glycol Immunoglobulin Con-
jugates: Resistance to Enzymatic Degradation. C. Cunningham-Run-
dles, Z. Zhuo, B. Grifilth

7. Stimulation of IL-2 Secretion and de novo Antibody after in vivo
Treatment with Polyethylene Glycol-Conjugated Human Recombi-
nant Interleukin-2. C. Cunningham-Rundles, L. Mayer, K. Kasbay

8. Adhesion Molecules in Intestinal Epithelium: Reduced Expression
in Active Inflammatory Bowel Disease. A. Panja, N. Chandswang, Y.
Li, L. Mayer

9. Regulation of Antigen Receptor Transcripts in Mast Cells by BUDR
and Interleukins. E. Siden

10. Do Normal and/or Diseased Intestinal Epithelial Cells Express
Superantigens? J. Aisenberg, D. Posnett, A. Pizzimenti, L. Mayer

11. Macrophage-Derived Mucus Secretagogue, a Product of Lamina
Propria Macrophages, Induces Mucus Release from Normal and In-
flammatory Bowel Disease Epithelium. J. Aisenberg, K. Sperber, S.
Itzkowitz, L. Mayer

12. RFLP at the Gamma T Cell Receptor Locus Found in Ulcerative
Colitis. A. Han, D. Posnett, L. Mayer

13. Identification of Monoclonal Antibodies which Recognize a Hu-
man B Cell Differentiation Factor. Y-D. Kuang, M. Cidon, L. Mayer

14. Protein Kinase A Activation by a Human B Cell Differentiation
Factor. R. Huang, L. Mayer

15. Response to a Human B Cell Differentiation Factor Results in the
Release of Intracellular Calcium. R. Huang, J. Cioffi, R. Kimberly, L.
Mayer*

16. Failure of Secretion of a Human B Cell Differentiation Factor by
Patients with Common Variable Immunodeficiency. K. Kazbay, C.
Cunningham-Rundles, L. Mayer*

Endocrinology

17. T-Cell Receptor V Gene Usage in Autoimmune Thyroid Disease:
Direct Assessment by Thyroid Aspiration. T. Davies, E. Concepcion, A.
Ben-Nun, P. Graves, G. Tarjan

18. Use of Transgenic Mice Expressing Human Neurofilament-M as
Donors for Preoptic Area Brain Grafts in Hypogonadal Mice to Study
Host-Graft Interactions. M. Gibson, V. Friedrich, G. Elder, R. Laz-
zarini, A-J. Silverman

19. Truncated TSH Receptor mRNA Variants in Normal Human Thy-
roid Tissue. P. Graves, Y. Tomer, T. Davies

20. Detection of Estrogen Receptor mRNA in Human Benign Prostat-
ic Hyperplasia. A. Kirschenbaum, I. Erenberg, B. Schachter, M. Ren,
A. Levine

21. T Cell Receptor V Gene Analysis of Human T Cell Lines and
Clones Reactive to Human Thyroid Peroxidase. A. Martin, E. Concep-
cion, N. Matsuoka, R. Magnusson, P. Graves

22. A Continuous Quality-Improvement Program for Diabetes in a
Large Metropolitan Hospital. S. Roman, P. Linekin, A. Stagnaro-
Green

23. Thiazolidinediones Ameliorate Glucocorticoid-Induced Insulin
Resistance in Rat Skeletal Muscle without Changing GLUT4 Glucose
Transporter Protein Content. S. Weinstein, A. Holand, E. O'Boyle, R.
Haber

24. Tumor Derived IL-2 Prevents the Induction of Immune Unrespon-
siveness by Murine Solid Tumor Cells. K. Zier, S. Salvadori, B.
Gansbacher

25. Increased Incidence of Postpartum Thyroid Dysfunction in
Women with Type I Diabetes. M. Alvarez-Marfany, S. Roman, A. Drex-
ler, C. Robertson, A. Stagnaro-Green

26. Restricted Survival of Intrathyroidal T Cells in Reconstituted Hu-
man Thyroid "Organoids" of SCID Mice as Assessed by T Cell Recep-
tor V Gene Analysis. N. Matsuoka, A. Martin, E. Concepcion, P. Unger,
L. Shultz, T. Davies*

27. T Cell Receptor V p Chain Gene Use in Autoimmune Thyroiditis of
NOD Mice. N. Matsuoaka, N. Bernard, L. Concepcion, P. Graves, A.
Ben-Nun, T. Davies

28. Liquid Hybridization Analysis for Human TSH Receptor mRNA.

Y. Tomer, P. Graves, T. Davies

29. NMA Stimulates Quiescent Opioidergic Afferents which Modulate
GNRH Release in the Intact Male Mouse. G. Miller, M. Gibson

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

247

248

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

Gastroenterology

30. Food Augments First Pass Metabolism of Alcohol by Promoting
Gastric Retention. R. Gentry, R. Lim, V. Zhuiin, C. Lieber

31. Hepatic Iron Metabolism in Chronic Hepatitis C. B. Hudes, T.
Fabry, F. Klion

32. Mucin Gene Expression in Colon Cancer. S. Ogata, H. Uehara, S.
Itzkowitz

33. Serum Lipase Levels in Abdominal Pain of Nonpancreatic Origin
vs. Acute Pancreatitis. N. Roditis, V. Gumaste, D. Mehta, P. Dave

34. A Statistical Analysis of Carcinoid Lab and Clinical Correlations.

R. Warner, S. Mani, K. Katz

35. Endoscopic Sphincterotomy Techniques and Complications. S. Co-
hen, F. Kasmin, F. Rutkovsky, J. Siegel

36. Minor Papilla Sphincterotomy in Pancreas Divisum-Complica-
tions and Response. F. Rutkovsky, S. Cohen, F. Kasmin, J. Siegel

37. Sialosyl-Tn Antigen Is Expressed Widely in Nondysplastic Mucosa
of Ulcerative Colitis and Crohn's Disease Patients with Colon Cancer.

A. Marshall, N. Harpaz, J. McHugh, A. Kombluth, I. Gelernt, A. Chen,
S. Itzkowitz

38. Expression of Blood Group A Antigen in Colon Cancer Tissues Is
not Predictive of Patient Survival. A. Weiss, G. Slater, A. Sandler, C.
Frissora-Rodeo, C. Bodian, A. Chen, S. Itzkowitz

General Medicine

39. Leukocyte Scanning with Indium In-Ill Oxyquinoline Is Diagnos-
tically Superior to Magnetic Resonance Imaging for Clinically Unsus-
pected Osteomyelitis in Diabetic Food Ulcers. L. Newman, J. Waller,
C. Palestro, G. Hermann, M. Klein, M. Schwartz, E. Harrington, M.
Harrington, S. Roman, A. Stagnaro-Green

40. Reduced Bone Mass in Women with Premenstrual Syndrome. S.

Thys-Jacobs, M. Silverton, J. Alvir, P. Paddison, M. Rico, S. Goldsmith

Geriatrics

41. Significant Premenopausal Bone Loss in White and Black
Women: a Longitudinal Study. D. Meier, M. Luckey, S. Wallenstein,
R. Lapinski

42. Physiological Consequences of Chronically Increased Vasopressin
Secretion in Transgenic Mice. M. Miller, S. Kawabata, B. Kent, M.
Wiltshire-Clement, J. Gordon

43. Utility of New York State Health Care Proxy Law for a Geriatric
Clinic Population. K. Mertz, D. Meier, B. Paris, M. Mulvihill, G. Gold,
A. Seckler, M. Beach, M. Weiss

Hematology

44. Molecular Cloning of a Mouse Locus Disrupted by Insertional
Mutagenesis Resulting in Fetal Anemia. L. Kiang, J. Wei, J. Krein-
dler, J. Gordon, L. Isola

45. Chromosomal Assignments of Four Hematopoietic Genes Using
Fluorescent in situ Hybridization Methodology. V. Najfeld, A. Ry-
bicki, J. Kehrl

46. Hematopoietic Progenitor Cell Transplantation at The Mount
Sinai Hospital, 1990-1992. E. Scigliano, A. Abramowitz, G. Ross, J.
Lipton, D. Tabrizi, L. Mandell, B. Shank, S. Fruchtman

47. Cell Factor Amplifies Normal Adult, Newborn, and Sickle Eryth-
ropoiesis in Liquid Cultures. R. Weinberg, J. Thomson, R. Lao, G.
Chen, B. Alter

48. Stem Cell Factor Stimulates in vitro Growth of Erythroid Progen-
itor Cells from HIV-t- Patients. R. Weinberg, E. Chusid, Y. Galperin,
T. Cheung, H. Sacks

49. Chromosomal Integration of HIV Determines its Cytopathic Ef-
fects upon T4 Lymphocytes. A. Melnick, I. Yu, L. Kiang, A. Atkin, L.
Isola

Infectious Diseases

50. Comparison of Clostridium difficile Diarrhea in Hospitalized HIV
Seropositive and HIV Seronegative Patients. A. Gurtman, S. Szabo, D.
Rose, H. Mayer, H. Sacks

51. Comparative Cross-over Study of a Needleless Heparin Lock Sys-
tem vs. a Conventional Heparin Lock: Impact on Complications,
Sharps Injuries, and Cost. M. Mendelson, L. Short, C. Schecter, B.
Meyers, M. Rodriguez, S. Cohen, J. Lozada, G. Button, H. Sacks, M.
DeCambre, P. Landrigan, S. Hirschman

52. Relationship of Antibiotic Usage and Bacterial Resistance: A Pro-
spective In-Hospital Study. B. Meyers, A. Gurtman, T. Abemathy, M.
Mendelson, C. Tejero, H. Sacks, S. Hirschman

53. Hyperglycemia during Pentamadine Treatment of Pneumocystis
carina Pneumonia in AIDS Patients. S. Zalasin, C. Scher, G. Fulop

54. Cost Effectiveness of Ganciclovir versus Foscarnet for Cytomeg-
alovirus Retinitis in Patients with AIDS. T. Cheung, M. Fahs, H. Sacks

Liver Diseases

55. Bile Duct Changes of Chronic Hepatitis C Are not Associated with
Abnormalities of Alkaline Phosphatase. N. Bach, F. Schaffner, S.
Thung

.56. Does Silicone Exposure Predispose to the Development of Primary
Biliary Cirrhosis? N. Bach, F. Schaffner, P. Berk

57. Prevalence of Primary Biliary Cirrhosis in Family Members of
Affected Patients. N. Bach, F. Schaffner

58. Hepatitis C Virus RNA Detection by Polymerase Chain Reaction
Correlates with Alanine Aminotransferase Changes. H. Bodenheimer,
S. Finkelstein, S. Cortez, R. Sayegh, S. Christensen, P. Swalsky

59. Retransplantation in Hepatitis B: A Multicenter Experience. J.

Crippin, S. Carlen, A. Borcich, H. Bodenheimer

60. Depletion of Carotenoids and Tocopherols in Liver Diseases. M.

Leo, A. Rosman, S. Ahmed, J. Lasker, C. Lieber

61. Alterations in Iron Uptake from Transferrin by Rat Hepatocytes
Isolated Following Endotoxin Administration: Time Course, Specific-
ity, and Concentration Dependence. L. Loney, H. Zhang, P. Slott, J.
Rand, B. Potter

62. Treatment of Decompensated Chronic Hepatitis B with a Titrat-
able. Low Dose Regimen of Recombinant Interferon Alfa-2b. R. Per-
rillo, C. Tamburro, F. Regenstein, L. Balart, H. Bodenheimer, E. Schiff,
M. Silva, J. Albrecht, C. Bodicky, C. Campbell, B. Miller, B. Taylor, C.
Brodcur, K. Roach

63. Effects of Endotoxins (LPS) on Iron Uptake by Cultured Rat Liver
Cells. B. Potter, L. Loney, H. Zhang

64. Fatty-Acid Uptake by the Perfused Liver Is not Facilitated by
Albumin. D. Sorrentino, K. Van Ness, D. Stump, P. Berk

65. Palmitate Competitively Inhibits Oleate Uptake in the Intact
Liver. D. Sorrentino, K. Van Ness, D. Stump, P. Berk

66. Characterization of Two Distinct Components of Hepatic Oleate
Uptake. D. Stump, R. Nunes, D. Sorrentino, L. Isola, P. Berk

67. Effects of Ethanol on Long-Chain Fatty-Acid Uptake by Isolated
Rat Hepatocytes. K. Wolfe, Z, Zheng, B. Potter

68. The Inner Nuclear Membrane Lamin B Receptor Is Differentially
Phosphorylated during the Cell Cycle and Is a Substrate for A
p34crfc2-Type Protein Kinase during Mitosis. H. Worman, N. Segil, G.
Blobel, J-C. Courvalin

69. Efficacy of Hepatitis B Vaccine in Patients Undergoing Ortho-
topic Liver Transplantation. J. Bemer, M. Kadian, J. Post, A. Borcich

Vol. 60 No. 3

ABSTRACTS

249

70. Structural Organization of a Human Nuclear Lamin Gene. F. Lin,
H. Worman

71. Ethanol Metabolism in Cultured Hepatocytes from Alcohol-fed
Rats. J-Z. Ni, B. Potter

72. Mechanism of Induction of Increased Toxicity to Carbon Tetra-
chloride (CCI4) by Rat Hepatocytes in the Presence of Ethanol. J-Z.
Ni. B. Potter

73. Autoantibodies Against Nuclear Envelope Proteins in Primary Bil-
iary Cirrhosis. R. Nickowitz, H. Worman

74. HCV RNA in Liver Allografts: Correlation with Histopathology. C.

Noyer, M. Fried, S. Thung, M. Shindo, N. Theise, M. Schwartz, T.
Borcich, C. Miller, A. DiBisceglie

75. First Pass Metabolism of Alcohol: Absence of Diurnal Variation
and Inhibition by Cimetidine after an Evening Meal. R. Sharma, R.
Gentry, R. Lim, C. Lieber

76. Effects of Ethanol Feeding and Disulfiram Treatment of Ferritin
Metabolism. H. Zhang, B. Potter

77. Further Studies on Ferritin Uptake by Rat Hepatocytes. H. Zhang,
B. Potter

78. Contribution of the Stomach to First Pass Metabolism of Ethanol
in Deermice. V. Zhulin, R. Gentry, R. Lim, E. Baraona, C. Lieber

79. Survival After Orthotopic Liver Transplantation in Alcoholics Is
not Different from other Groups. N. Freedman, J. Kamean, T. Fabry,
0. Butchma, W. Dewell, C. Miller, M. Schwartz, E. Katz, F. Klion, A.
Borcich

80. Quality of Life after Liver Transplantation in Alcoholic vs. Non-
Alcoholic Recipients. J. Kamean, N. Freedman, M. Schwartz, C.
Miller, T. Fabry, J. Dewell, A. Borcich

81. Underreporting of Recidivism after Orthotopic Liver Transplanta-
tion in Alcoholic Patients. J. Kamean, N. Freedman, T. Fabry, C.
Miller, M. Schwartz, A. Borcich

82. The Amino Terminal Domain of the Lamin B Receptor Is a Nu-
clear Envelope Targeting Signal. B. Soullam, H. Worman*

Molecular Medicine

83. Identification and Characterization of pRASM 20, a Novel Growth
Related Gene in Vascular Smooth Muscle. S. Wax, C-L. Rosenfield, M.
Taubman*

nosis of Disseminated Candidiasis. J. Roboz, Q. Yu, L-H. Ma, J. Hol-
land

90. RB Protein Translocation and Protein Kinase (' Stimulation Are
Associated with Cell Growth Inhibition. V. Rogalsky, G. Todorov, D.
Moran, T. Den, T. Ohnuma

91. Cloning and Sequencing of MMTV ENV Gene-Like Sequences
from Human Breast Cancer. Y. Wang, B. Pogo, J. Holland

92. The Effects of lonophores and Noeodazole on Multidrug Resis-
tance. L-T. Wu, H Arkin, J. Holland, T. Ohnuma

93. Cleavage of Human Multidrug Resistance mRNA by a Hammer-
head Ribozyme. H. Kobayashi, T. Dorai, J. Holland, T. Ohnuma*

94. Relationship Between Vincristine Binding to Cellular Targets and
its Cell Kill Effects. H. Kobayashi, T. Ohnuma, J. Holland

Pulmonary

95. Patterns of Asthma Admissions to a Metropolitan Hospital 1990.

M. Cabrera, N. Rienzi, E. Schachter

96. Intra-Allograft Production and Systemic Release of TNF-ALPHA:
Detection upon Reperfusion. T. Kalb, M. Walter, K. Acarli, K. Fuku-
zawa, L. Mayer, M. Schwartz, C. Miller

97. Alteration in Exercise Performance with Thoracic Strapping — A
Model for Chest Wall Restriction. A. Miller, M. Sloane, A. Bhuptani,
L. Brown, A. Teirstein

98. Comparison of Estimated versus Calculated V,>/Vt- during Cardio-
pulmonary Exercise Testing. A. Miller, M. Zimmerman, A. Bhuptani,
M. Sloane, L. Brown, A. Teirstein

99. Mixed Venous O2 Saturation: Measured by Co-oximetry versus
Calculated from PV02. D. Nierman, C. Schecter

100. Variation in Asthma Admission Rates in New York City. V. De-

Palo, P. Mayo, M. Rosen

101. Unopposed Stimulation of Muscarinic Receptors Causes Bronchi-
al Hyper-responsiveness after Cervical Spinal Cord Injury. P. Dic-
pinigaitis, P. AlmenofT, A. Absgarten, A. Spungen, W. Bauman

102. The Changing Spectrum of Pulmonary Complications of HIV
Infection. J. Ruzi, M. Rosen

103. Airway Reactivity in Fabry's Disease. M. Zimmerman, L. Brown,
A. Miller, A. Teirstein, R. Desnick

Oncology /Neoplastics

84. Evidence of a Pathologic Role for Polypeptide and Endocrine Tu-
mor Markers in Patients with Unresectable and Metastatic Pancreatic
Cancer. H. Bruckner, A. Gattani, M. Chesser, J, Mandeli, L. Farber, G.
DiGiovanni

85. Increased in vitro Uptake of Doxorubicin in Human Breast Car-
cinoma MCF-7 Cells following Exposure to a Somatostatin Analogue
(SMS 201-995). Further Evidence of Polypeptide Hormones as a New
Class of Biochemical Modulators for Use in Cancer Chemotherapy. H.
Bruckner, J. Roboz, J. Lee, R. Erlich, T. Ohnuma

86. Surgical Resection Following Combined-Modality Therapy for Un-
resectable Stage II-III Pancreatic Cancer. H. Bruckner, A. Cooper-
man, H. Snady, M. Chesser, A. Gattani, M. Sung, J. Mandeli

87. A Nude Mouse Model for the Therapy of Mesothelioma. P. Cha-
hinian, L. Szrajer, H. Gluck, J. Holland

88. New Survival Standards and Goals for Pancreatic Cancer Therapy
Trials. A. Gattani, J. Morris, A. Kamthan, J. Mandeli, J. Dalton, M.
Chesser, H. Snady, A. Cooperman, J. Siegel, H. Bruckner

89. Progress in the Use of Serum D/L Arabinitol Ratios for the Diag-

Renal

104. Okadaic Acid, A Specific Inhibitor of Protein Phosphatase, Stim-
ulates NaK2Cl Cotransport in Cultured Mouse Medullary Thick As-
cending Limb Cells. D. Kaji

Rheumatology

105. Fulminant Myocarditis as a Complication of Scleroderma Pa-
tients with Myositis. L. Kerr, H. Spiera

106. Scleroderma Following Silicone Implantation: a Cumulative Ex-
perience of Twelve Cases. H. Spiera, L. Kerr

107. An Update on the Occurrence of Penicillinase-Producing Organ-
isms Causing Gonococcal Arthritis in New York City. M. Omstein

108. The Use of Parenteral ACTH for Acute Crystal-Induced Synovitis
in Patients with Multiple Medical Problems. J. Ritter, L. Kerr, J.
Valeriano-Marcet, H. Spiera

Abstracts

These abstracts summarize research in ophthalmology by investigators working at Mount
Sinai School of Medicine (CUNY) in New York City. They are part of the 1993 Association for
Research in Vision and Ophthalmology abstracts and are published here concurrently with
publication in Investigative Ophthalmology.

Abstracts in Ophthalmology from
Mount Sinai Medical Center 1993

Alpha-Adrenergic Antagonists: Effects on Aqueous Humor Dynamics
in the Rabbit and Monkey Eye. T. Taniguchit, S. M. Podost and T. W.
Mittaqi^*. Depts. of Ophthalmology^ and Pharmacology*, Mount Sinai
School of Medicine, New York, NY.

Purpose. Some a; -adrenergic antagonists such as prazosin are modest
ocular hypotensive agents, but their mechanism(s) of action is uncer-
tain. Recent cloning studies show three mammalian receptor sub-
types, A, B and C. Prazosin has highest affinity at A and B
receptors. As yet there are no highly subtype-selective antagonists.
We have investigated the aj-antagonist BE2254 (BE), which has high
affinity binding sites in rabbit ciliary processes consistent with C
receptors. Methods and Results. Topical BE (0.2-0.5%) lowered lOP in
rabbit and in normal and glaucomatous monkey eyes by 4-8 mmHg
from 1 hr to 6 hours after treatment. At 1 hr after 0.5% BE, rabbit
outflow facility measured tonographically (untreated 0.28 ± 0.07
(S.D.); treated 0.29 ± 0.07, n = 8) or by two-level constant pressure
perfusion (untreated 0.18 ± 0.010; treated 0.19 ± 0.011) was un-
changed. Uveoscleral outflow, determined at 1-1.5 hrs post-BE by flu-
orescein-dextran perfusion in rabbit eyes, showed no significant drug
effect (saline- treated 0.29 ± 0.05, untreated eye 0.27 ± 0.10, drug-
treated eye 0.31 ± 0.08 |xl/min at 15 mmHg). However, aqueous humor
formation measured fluorophotometrically was significantly (p < 0.05)
decreased in rabbit eyes over a 3 hr period after 0.5% BE treatment
(baseline 2.47 ± 0.60, treated 1.85 ± 0.53 fxl/min). Conclusions. These
results suggest that aqueous humor formation by ciliary processes is
modulated by C adrenergic receptors in addition to the well-known
responses to a2 and ^2 adrenergic receptor drugs.

Antidepressants and an Hj Blocker Reduce CI Transport in Frog Cor-
neal Epithelium by Decreasing K Permeability. Aldo Zamudio and
Oscar A. Candia. Mount Sinai School of Medicine, New York, NY.

The cellular mechanisms involved in the effects of antidepressants
such as fluoxetine (F) (which inhibits serotonin uptake at the synaptic
cleft) and nortriptyline (N) remain poorly understood. Unrelated to
this, we found that the histamine receptor (Hi) blocker diphenhy-
dramine (D), inhibits, at 10 " ^ M, 01 transport across the isolated frog
corneal epithelium. Because tricyclic antidepressants and fluoxetine
bind to an intracellular histamine receptor (Hjc) (Brandes LJ, et al.
Cancer Research 1992; 52:3796-3800 ), we tested the effects of fluox-
etine and nortriptyline (at 10"^ M) on CI transport. Corneas were
mounted in an Ussing-type chamber and bathed with Ringer's. The
antidepressants reduced the Cl-originated Igc by 50% while increasing
the resistance by 30% . The effects were only observed from the apical
side and could not be blocked by a-adrenergic antagonists. Microelec-
trode impalements showed a 20 mV cellular depolarization, consistent
with a decreased basolateral K permeability. Corneas were also bathed
with a high-K, Cl-free, Na-free solution on the apical side which was
permeabilized with amphotericin B. Under this condition Igc repre-
sents a K current across the basolateral membrane. This current was
reduced by D (66%), F (46%) and N (55%). It seems that these drugs

exert their cellular effects by interacting with apically-accessible Hi or
Hic receptors which are linked to the regulation of K channels.
Supported by NIH grants EY00160 and EY01867.

Choroidal Vascular Tone-Potential Biochemical-Pharmacological
Mechanisms. K. G. Schmidt, 0. Geyer, and T. W. Mittaq*. Depart-
ments of Ophthalmology & Pharmacology*, Mount Sinai School of
Medicine, New York, NY.

The choroid, a low resistance vascular structure carrying 85% of the
ocular blood flow, has important functions including providing nour-
ishment to and removal of potential toxic waste products from the
adjacent non-vascularized retina and optic disc region. Few studies
exist examining the biochemical or pharmacological mechanisms for
maintenance of choroidal vascular tone. Therefore, we have studied
some basic characteristics of the adenylyl (AC) and guanylyl cyclase
(GO) systems in the bovine choroid. Compared to respective baseline
measurements ( = 100 ± SEM%), AC responses were increased (p <
0.05) by forskolin (FSK, 477 ± 59%), fluoroaluminate (AIF4, 360 ±
10.3%), isoproterenol (ISO, 129 ± 5.5%), vasoactive-intestinal peptide
(VIP, 132 ± 6.1%), calmodulin (CAM) + Mn^" (196 ± 69%), and dose-
dependently by PGE2 (up to 162 ± 3.6%). The antagonist drug Calm-
idazolium inhibited the CAM dependent increase but also blocked
basal activity by 47 ± 2.0% at 50 |xM without affecting the FSK re-
sponse. Other CAM blockers (TFP, W5) produced similar results, but
were not completely selective for CAM-cyclase. GDP (3 S did not affect
AC responses to FSK, ISO, and ALF4, but decreased the response to
PGE2. The adenosine agonist NECA did not influence FSK or ISO
responses, and did not activate AC by itself. In the GC system activity
was increased (p < 0.05) by CAM + Mn^^ (239 ± 27%), by atrial
natriuretic peptide (up to 143 ± 1.4%) and sodium nitroprusside (up to
179 ± 1.6% at 10 (iM).

Conclusions. These results show that choroidal tissue has signifi-
cant activities of the adenylyl cyclase and guanylyl cyclase second mes-
senger systems that may control the degree of relaxation in this vas-
cular tissue.

Supported by Deutsche Forschungsgemeinschaft, Sch 870 1/1; Cas-
troviejo Foundation; and USPHS (NEI) grant EY02619.

Control and Regulation of CI Secretion by the Frog Corneal Epithe-
lium. Oscar A. Candia and Aldo Zamudio. Mount Sinai School of Med-
icine, New York, NY.

CI secretion is controlled by at least five elements of the transporting
epithelial cell layer: the Na-K pump, Na-Cl (K) cotransporter, basolat-
eral K and Na permeabilities and apical CI permeability. To further
elucidate the stoichiometry and relationship between the rates of CI,
Na and K transport the following experiments were made. Isolated frog
corneas were placed in an Ussing-type chamber and bathed with a
cell-like solution on the apical side and normal Ringer's on the baso-
lateral side. Isc. ^''Cl fluxes and *''Rb fluxes (as a K label) were mea-
sured simultaneously in either direction (S = stroma; T = tear) under

250

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

Vol. 60 No. 3

ABSTRACTS

251

control conditions and after sequential treatment with amphotericin B
(to permeabilize the apical membrane to ions), quinidine, ouabain and
bumetanide. Jci st was reduced by quinidine (due to cell depolariza-
tion), ouabain and bumetanide. Jci ts was reduced only by bumetanide.
jRb.sT was reduced by quinidine (45%), ouabain (36%) and bumetanide
(19%). Ts was reduced by quinidine and bumetanide. The inhibition
by the diuretic of Jci.st was twice its effect on Jph sx- It 's concluded
that: (a) CI movement across the basolateral membrane is bidirec-
tional, limited to the cotransporter, and coupled to Na and K; (b) apical
[CD in the range 0-100 mM has no influence on the rate of CI secretion
which is mainly controlled by the elements on the basolateral mem-
brane and regulated by apical CI permeability.

Supported by NIH grants EY00160 and EY01867.

Corneal Topography in Keratoconus Contact Lens Fitting. A. N. Co-
hen, M. J. Dunn, A. Bartolomei, and P. A. Asbell. Mount Sinai Medical
Center, New York, NY.

Purpose. Traditionally contact lens fittmg utilizes central keratometry
readings to estimate the base curve for initial trial lens fitting. Best fit
is eventually achieved through subsequent trial and error. This be-
comes even more important in keratoconus patients where there is a
significant amount of asymetrical astigmatism. Methods. We examined
20 eyes of 12 patients with keratoconus which were successfully fitted
with standard rigid gas permeable contact lenses to determine if there
was a statistically significant correlation between the base curve (B.C.)
selected and the keratometry readings obtained by corneal topography.
Results. The average B.C. was 46.8 ± 6. ID. The average keratometry
readings were at 3mm fiat 47.0 ± 7.0D and steep 50.7 ± 9.0D; 5mm flat
45.35 ± 5.93D and steep 49.29 ± 7.82D; at 7mm flat 45.75 ± 4.76D and
steep 48.18 ± 7. 2D. There was no statistically significant correlation
between the base curve and standard keratometry readings. Looking at
the corneal topography we found that the area that best correlated with
the B.C. was the transition zone where the topography changed from
the steep area of the cone to the area of the flatter, more normal cornea.
Keratometry readings at the 3, 5, or 7 mm diameter were not good
predictors of the base curve used for best fit. Conclusion. No statisti-
cally significant correlation was found between the base curve selected
and the keratometry readings. Corneal topography demonstrated a
transition zone which was a good indicator for selecting the base curve.
A computerized program is now being developed to automatically pick
a base curve and simulate fluorescein patterns on the computer.
Supported by NEI EY01867.

The Effect of Brimonidine Tartrate in Glaucoma Patients on Maximal
Medical Therapy. J. B. Serle'", S. M. Podos'^', G. P. Abundo'^', R.
Ritch'^', A. L. Coleman'^', W. C. Panek'^', C. E. Mantras"". Mount
Sinai Medical Center'", and New York Eye & Ear Infirmary'^', New
York, NY, Jules Stein Eye Institute'^', Los Angeles, CA, Allergan,
Inc."*', Irvine, CA.

Purpose. We evaluated the efficacy of the topical alpha2-agonist, bri-
monidine tartrate, in delaying surgery in patients in whom present
maximally tolerated glaucoma therapy was insufficient. Methods. In
an open-labeled, non-comparative study, patients were treated with
brimonidine 0.2%, b i d., in one or both eyes, in addition to existing
maximally tolerated glaucoma medications. Study medication was ti-
trated up to 0.5% brimonidine if necessary. Follow-up examinations
were scheduled at one day, one week, 1, 2, and 3 months, and every
three months for up to one year. Results. Of the 50 patients (74 eyes)
enrolled, 47% (35 eyes) are still continuing therapy with 0.2% brimoni-
dine. Twenty-two patients have undergone therapy with 0.2% brimoni-
dine for at least three months. Mean follow-up time for all patients
with 0.2%. brimonidine was 106 days (min = 1, max = 327). The
medication was titrated to the 0.5%. concentration in 17 eyes (24% ). The
probability of successful treatment with 0.2% brimonidine for three
months is 70%, and is 50% for more than six months. Conclusions.
Brimonidine tartrate, used as adjunctive therapy, was effective in de-
laying surgery in patients on maximally tolerated medications.
Supported by Allergan, Inc./E.

The Effect of Flurbiprofen (FL) on the Histopatholoify of Pigmented
Rabbit Eyes after Contact Transscleral Nd:YAG Laser Cyclophotoco-
agulation (CYC). A. H. Friedman, K. G. Schmidt, C. B. Camras, and
S. M. Podos. Ophthalmology, Mount Sinai School of Medicine, New
York, NY.

The correlation of histopathologic changes after CYC with reduction of
intraocular pressure (lOP) is unclear. Destruction of the secretory cil-
iary epithelium or liberation of prostaglandins may account for CYC's
action. CYC (30 applications 360°, each at 3 W and 0.5 sec and centered
1.5 mm from the limbus) was performed in one eye each of 14 pig-
mented rabbits, and a sham procedure in the contralateral eye. In an
additional 15 pigmented rabbits, both eyes underwent CYC, one eye
treated with FL and the other with vehicle. The drops were applied
every 30 min for 2 hrs before CYC, every 2 hrs for 24 hrs, and continu-
ing every 6 hrs thereafter. lOPs were measured and eyes were histo-
logically examined on day 1 and 17 following CYC. On day 1, in the
group not receiving FL, the CYC-treated eyes (n = 8) showed coagu-
lation necrosis of the inner sclera with vaporization of ciliary processes.
lOP was reduced (p < 0.0001) by 12.0 ± 0.7 mmHg(mean ± SEM; n =
14) compared to baseline values. On day 17, these eyes (n = 6) showed
atrophic areas in the inner sclera with overlying fibrous plaques and
their lOP had returned to baseline values. On day 1, the eyes treated
with CYC and FL (n = 9) displayed coagulation necrosis of the inner
sclera. The overlying ciliary body was vaporized. lOP was reduced (p <
0.0001) by 12.8 ± 0.6 mmHg (n = 15) compared to baseline. On day 17,
these eyes (n = 6) displayed a normal outer sclera. Fibrous plaques
were present at the sites of ciliary body vaporization. lOP had returned
to baseline values. There were no qualitative histopathological or lOP
differences in eyes treated with FL or not.

Deutsche Forschungsgemeinschaft, Sch 870 1/1; EY07865 and
EY01867.

Effect of Oxymetazoline on Aqueous Humor Dynamics and Ocular
Blood Flow in Monkeys and Rabbits. S. M Podos, R-F Wang, P-Y Lee,
T. Taniguchi, B. Becker*, J. B. Serle, I. W. Mattag. Departments of
Ophthalmology, Mount Sinai School of Medicine, New York, NY and
Washington University School of Medicine,* St. Louis, MO.

An investigation was carried out to determine the mechanism by which
oxymetazoline, an a2-adrenergic agonist, reduces intraocular pressure
(lOP) when topically applied. Tonographic outflow facility (C), fluoro-
photometric aqueous humor flow (F), uveoscleral outflow (Fu) deter-
mined by a perfusion technique with FITC-dextran, and ocular peak
pulse volume (PPV) determined by the Langham Ocular Blood Flow
System were measured before and after therapy in the eyes of cyno-
molgus monkeys and albino rabbits. In 8 normal monkeys, C was un-
changed (p > 0.10) at 2 hrs after single dose administration of 0.5%
oxymetazoline, and F was significantly (p < 0.001) reduced by 37%
over 4 hrs following 1% oxymetazoline [1.02 ± 0.05 (xl/min (SEM)
(treated) vs 1.62 ± 0.19 jj,l/min (contralateral)]. In 8 rabbits, Fu was
significantly (p < 0.005) increased by 56% at 3 hrs after 0.1% oxymeta-
zoline application [0.28 ± 0.02 (xl/min (treated) vs 0.18 ± 0.03 (xl/min
(contralateral)]. 0.5% oxymetazoline did not alter PPV in the eyes of 6
normal and 5 glaucomatous monkeys.

The results suggest that oxymetazoline reduces lOP by decreasing
aqueous humor flow and by increasing uveoscleral outflow.

The Effect of Prozac on Potassium and Sodium (Na) Currents in Lens
and Corneal Epithelia. J. L. Rae,* A. J. Rich, and O. A. Candia'''. Depts
of Physiology and Biophysics and Ophthalmology, Mayo Foundation,
Rochester, MN* and Depts. of Ophthalmology and Physiology, Mount
Sinai School of Medicinet, New York, NY.

Prozac is a drug with antidepressive action in humans in doses of 0.3- 1
\jM. It also has blocking effects on at least 3 ion channels in lens and
corneal epithelium when used at higher concentrations (1.0-10.0 x
10 " ^ M). In rabbit corneal epithelium, Prozac reduces the highly stud-
ied whole cell potassium current and causes cell depolarization when it

252

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

is added to either a NaCl of KCl ringer bathing the cells. The current
reduction is substantial at all voltages between -70 to + 100 mV but
not complete at positive voltages even when used at 1 x 10 M.

In cultured human and bovine lens epithelium, Prozac substan-
tially reduces delayed rectifier potassium currents while having little
effect on inward rectifying and calcium activated potassium currents.
Again, 1 x 10 M Prozac is insufficient to totally block delayed rec-
tification at positive voltages. Cultured human lens epithelium also
expresses tetrodotoxin blockable sodium channels. These are also
blocked substantially by Prozac in concentrations as low as 1 x 10"*
M, a block which is only partially reversible.

While the effective concentrations of Prozac required to block oc-
ular potassium channels are higher than the therapeutic doses used in
humans, they are as low or lower than those of other commonly used
potassium channel blockers. Prozac may therefore prove to be a useful
tool for dissecting different potassium conductances in ocular epithelia
and elsewhere.

Supported by EY03282, EY06005, and EY00160 and EY01867.

The Effects of Isoproterenol and Forskolin on the Phagocytosis of
Rod Outer Segments by Retinal Pigment Epithelial Cells Are Inde-
pendent of cAMP and Protein Kinase A. Shoji Kuriyama' ' ', Michael O.
Hair Toshka A. Abrams' ' ', and Thomas W. Mittag''^'. Jules Stein Eye
Institute, UCLA School of Medicine, Los Angeles, CA'^', and Depart-
ment of Ophthalmology, The Mount Sinai Medical Center, New York,
NY '2'.

Purpose. We have studied the involvement of cAMP dependent second
messenger systems in the inhibition of ROS phagocytosis by isopro-
terenol (ISO) and forskolin (FSK) using two membrane permeant an-
alogs of cAMP, the Rp and Sp diastereomers of cyclic adenosine 3', 5'
monophosphothioate (cAMPS), and antisense oligodeoxynucleotides to
Gs protein a-subunit sequence (Gsa). Methods and Results. Rp-cAMPS
is a particularly potent competitive inhibitor of PKA, whereas Sp-
cAMPS is a potent activator of this enzyme. Sp-cAMPS showed a dose-
dependent inhibition of ROS phagocytosis (ID50 = 10 p.M), while Rp-
cAMPS had no effect on this process. Rp-cAMPS prevented the
inhibitory effect of Sp-cAMPS, but had no effect on the inhibition of
ROS phagocytosis induced by ISO and FSK. ISO plus FSK showed an
additive effect on phagocytosis, while ISO plus Sp-cAMPS and FSK
plus Sp-cAMPS did not. RPE cells were treated with antisense Gsa
oligodeoxynucleotides, after which the ability of ISO to inhibit phago-
cytosis was measured. The treated cells showed a greatly decreased
inhibition of phagocytosis (30% of control, untreated cells). This indi-
cates that Gs is directly associated with the inhibitory effect of ISO.
Conclusion. Our results suggest that ISO and FSK inhibit ROS phago-
cytosis by RPE cells through a cAMP-independent (PKA-independent)
pathway, as well as through a cAMP-dependent one. The cAMP-inde-
pendent pathway may be mediated directly through a G protein.

Supported by grants EY00046, EY00331 (M.O.H.) and EY02619,
EY07400 (TWM) from USPHS, by a grant from the National Retinitis
Pigmentosa Foundation Fighting Blindness Inc. (MOH) and by a Fight
For Sight grant PD92012 (SK).

Effects of Medetomidine (MED) on Intraocular Pressure (lOP), Pupil
Diameter and Uveoscleral Outflow in Rabbits. P-Y Lee, J. B. Serle,
and S. M. Podos. Department of Ophthalmology, Mount Sinai School of
Medicine, New York, NY.

Recent studies have demonstrated that MED, a selective alpha-2 ago-
nist, with affinity for alpha-2 receptors that is approximately three
times higher than clonidine, decreases lOP in rabbits. The mechanisms
which cause this lOP reduction are not known. To explore possible
mechanisms, lOP was measured by pneumatonometry, and uveoscleral
outflow (Fu) was determined by a perfusion technique with FITC-dex-
tran. Pupil diameter (PD) was measured with a clear millimeter ruler
in normal room light. lOP and PD measurements were taken at 0 hr,

0.5 hr, and hourly for a total of 6 hrs on one baseline day with vehicle
administration, and subsequently after 25 \lI of 0.1%, 0.5% or 1% MED
was applied to one eye and an equal volume of vehicle to the contra-
lateral eye in 6 rabbits. lOP was significantly (p < 0.05) reduced bi-
laterally and PD was significantly (p < 0.01) increased in the treated
eyes after 0.1% MED. The 0.5% and 1% concentrations of MED resulted
in (p < 0.02) lOP and PD effects of greater magnitude than 0.1%. Fu
was measured 2 hrs after 25 (il of 0.5%) MED was applied to one eye of
8 rabbits. MED produced a significant (p < 0.005) increase in Fu in the
treated eyes (0.40 ± 0.04 |xl/min) and contralateral control eyes (0.36 ±
0.04 ^.1/min) compared to 13 untreated eyes (0.23 ± 0.01 jjil/min).

Our results indicate that in rabbits MED reduces lOP bilaterally
following unilateral topical application. This lOP reduction is due in
part to an increase in Fu.

Effects of Topical Ethacrynic Acid (EY-105) Ointment on Intraocular
Pressure (lOP) in Glaucomatous Monkeys. R-F. Wang, J. B. Serle,
S. M. Podos, A. H. Neufeld*, R. Deschenes*. Department of Ophthal-
mology, Mount Sinai School of Medicine, New York, NY, and Telor
Ophthalmic Pharmaceuticals, Inc.*, Wobum, MA.

Ethacrynic acid, a sulfhydryl reactive drug, reduces lOP and increases
outflow facility in normal monkeys. In the present study, the effect of
topical EY-105 ointment on lOP was evaluated in 4 monkeys in which
glaucoma was induced by argon laser treatment of the trabecular
meshwork. lOP was measured daily at 0 hr, 0.5 hr and hourly for 7 hrs
during one baseline day, one vehicle-treated day, and five days of ther-
apy with 0.5%, 1.0%-, 1.5%^ or 2.5%^ EY-105 ointment applied once daily
at 9;00 a.m.

lOP was not different comparing baseline and vehicle-treated
days. All four concentrations of EY-105 ointment reduced lOP, al-
though the onset of lOP reduction did not occur until after the second
dose with any of the concentrations. The magnitude of the lOP reduc-
tion was dose-dependent. The mean maximum reduction in lOP was
8.5 ± 2.9 mmHg (SEM). A more pronounced reduction in lOP was
observed on the 5th day of treatment for each of the 4 concentrations.
An effect on lOP was observed for 24 hours after a single dose. Mild
conjunctival hyperemia and discharge appeared in some eyes treated
with the three highest concentrations. The corneas remained clear.
Neither aqueous flare nor anterior chamber cellular response were
apparent. EY-105 ointment may prove to be useful in glaucoma ther-
apy.

Indocyanine Green (ICG) Enhancement of Diode Laser Cyclophoto-
coagulation in Rabbits. Steven A. Odrich, Gregory L. Cowan, Francis
J. Wapner, Alan H. Friedman, Steven M. Podos. Mount Sinai School of
Medicine, New York, NY.

Purpose and Methods. In order to assess the effect of intravenous ICG
on diode laser contact cyclophotocoagulation (CPC), six pigmented rab-
bits received diode CPC under sedation and topical proparacine. Each
animal's right eye received no pre-treatment. Immediately before
treatment of their left eyes, animals were administered 1.0 ml of in-
travenous ICG (5 mg/ml). CPC was then performed in an identical
manner as in the contralateral eye. Each treatment, applied 1.0 mm
posterior to the limbus, consisted of a single 1.0 second pulse. At every
second treatment site output energy was increased by 0.1 J, starting
with a sub-threshold 0.1 J/pulse and peaking at the supra-threshold
level of 1.0 J/pulse. Threshold energies were defined as 0.6-0.7 J for
diode CPC in rabbits. Three animals were sacrificed at 24 hours, and
three at 72 hours using barbiturate overdose. All eyes were enucleated
and prepared for sectioning and staining. Results. Gross examination
revealed that two ICG-treated eyes had spontaneously ruptured at the
sites of laser application using threshold energy levels. No such find-
ings were seen at any energy level in eyes not pre-treated with ICG. On
histologic examination it was found that CPC lesions in the eye of an
animal pre-treated with ICG demonstrated more extensive coagulative

Vol. 60 No. 3

ABSTRACTS

253

necrosis than did lesions created with equal energies in the same an-
imal's control eye. Conclusions. Intravenous ICG appears to decrease
the threshold energy at which ablation of ocular tissues is achieved
using the diode laser, while increasing the risk of significant compli-
cations including destructive coagulative necrosis of non-targeted oc-
ular tissues, weakening of the eye wall, and frank rupturing of the
globe. Appropriate and safe energy levels must be established before
this technique is attempted in humans.

This postdoctoral fellowship has been awarded in honor of Elsie K.
Sloate by the Fight for Sight Research Division of the National Society
to Prevent Blindness.

Evaluation of Chemotaxis by Various Cytokines of la"^ Langerhans
Cells into the Central Cornea of Mice. S. P. Epstein\ M. S. Jan, and
P. A. Asbell'. Department of Ophthalmology, Mount Sinai Medical
Center, New York, NY.

Purpose. Under normal conditions the central cornea is devoid of la*
Langerhans cells (LC). We sought to determine which cytokines are
chemotactic for LC in the cornea. Methods. 50 BALB/c and 50 C57BL/6
mice were injected with a single intracomeal injection of 200 U recom-
binant (r) murine (m) IL-la, 0.004 U r mu IL-13, 250 U r mu IL-2, 1000
U r mu IL-6, 10 U natural (n) human TNF-a or 25 U n mu GM-CSF.
After a 2 hour (h) incubation, they were indirectly immunoperoxidase
stained with rat anti-mu la antibody. Results. Unstimulated corneas
were found to contain 0 to 3 la * LC. IL-ip, IL-6 or TNF-a induced the
migration of 8.3 ± 2 BALB/c and 11.8 ± 4 C57BL/6 la* LC/comea;
while IL-2 induced the migration of 25.5 ± 9 la * LC in BALB/c, but
only 12.2 ± 4 la * LC in C57BL/6 mouse corneas. Intracomeal injection
of IL-la, or GM-CSF were not chemotactic for LC and incubations as
long as overnight ( — 16 h) failed to elicit any greater chemotaxis than
those of 2 h for any of the cytokines studied. To determine whether the
effect of each cytokine was directly on LC or via the induction of a
common intermediary, we investigated whether Cyclosporine A (CsA)
could abrogate the cytokine-mediated chemotaxis of LC into the cor-
nea. In the presence of 5 jj.g CsA, both IL-2 and TNF-a remained che-
motactic for LC, while the chemotactic effects of IL-13 and IL-6 were
completely abrogated. Conclusions. We propose that IL-2 and TNF-a
are directly chemotactic for LC, whereas IL-ip and IL-6 act through
the induction of either IL-2, TNF-a or both.

Supported in part by NEI 5P30EY01867 and Research to Prevent
Blindness, Inc.

Evidence for a Bumetanide-Sensitive Na-K-CI Cotransporter in Cul-
tured Bovine Lens Epithelial Cells. Lawrence J. Alvarez, Aldo Zamu-
dio, J. Mario Wolosin, and Oscar A. Candia. Mount Sinai School of
Medicine, New York, NY.

Previous studies on the cells noted above indicated indirectly that a
Cl-uptake mechanism was present (ICER, 1992). This notion arose to
explain the finding that the Na-H exchanger was only activated by
hypertonicity if the cells were either pre-exposed to bumetanide or
pre-equilibrated under Cl-free conditions. Using '^^Rb as a label for K
recent experiments have demonstrated directly the presence of the in-
ferred transport activity. Epithelial monolayers (10 cm^) were incu-
bated in a HEPES-buffered physiological solution in the presence of the
label for 10 min. The counts were extracted in TCA and measured in a
gamma detector; protein was digested in NaOH for measurement by
the Lowry assay. An influx rate of 1.51 ± 0.43 |xEq/mg ■ hr (±SD; n =
13) was calculated. The bumetanide-sensitive component represented
43 ± 7% (n = 6) of the influx. The combination of bumetamide plus
ouabain inhibited 98 ± 2% (n = 6) of the uptake. In Cl-free solutions,
bumetanide did not affect the baseline uptake (1.03 ± 0.27; n = 6),
which was inhibited by 97 ± 3% upon ouabain addition (n = 4). The
small influx in the presence of the 2 inhibitors indicated little passive
movement of label via K channels, suggesting a relatively depolarized
cell system. Preliminary microelectrode impalements provided values
of about - 25 mV, a level consistent with the literature for the intact

bovine lens. It is suggested that the bumetanide-sensitive pathway also
exists in the lens.

Supported by NIH grants EY01867 and EY00160.

Focal ERG Pha.se-Lag in Diabetic Macular Edema. S E Brodie,"^'
D. E. Sperber,'^' and M. Hope-Ross.'^' Mount Sinai School of Medi-
cine,'" Manhattan Eye, Ear & Throat Hospital,"" New York, NY.

Purpose. We recorded focal electroretinograms (FERG) from patients
with diabetic macular edema in an attempt to confirm previous reports
of substantial phase lags with normal FERG amplitude in such pa-
tients. Methods. FERG recordings were obtained using the Doran In-
struments Maculoscope stimulator ophthalmoscope, which images a
five-degree flickering central spot on the macula under ophthalmo-
scopic control, surrounded by a brighter steady annulus to desensitize
the more peripheral retina to scattered light. Patients were selected
from the retina clinic at Manhattan Eye, Ear & Throat Hospital. Mac-
ular edema was confirmed by fluorescein angiography. Patients who
had undergone prior focal laser treatment were excluded. Informed
consent was obtained from all participants. Results. FERG recordings
in our patients demonstrated phase lags of 120 degrees or greater,
compared with an upper limit of normal of 95 to 100 degrees. Conclu-
sions. The pronounced phase-lag of the FERG seen in diabetic macular
edema appears to be a more robust abnormality than the small reduc-
tions in Ganzfeld ERG amplitude previously reported in this patient
population.

Supported by grants from Research to Prevent Blindness, the
LuEsther Mertz Retina Research Fund, and the Manhattan Eye, Ear &
Throat Hospital Foundation.

Identification of a Novel WGA-Binding Integral Membrane Protein
Associated with Corneal Epithelial Cell Stratification. Gabriel J. G.
Hou and J. Mario Wolosin. Mount Sinai School of Medicine, New York,
NY.

Previous studies have shown that all intrastratal suprabasal (SB) cells
of the rabbit corneal epithelium, when exposed to the outer corneal
surface by the digitonin-induced exfoliation of overlying cell layers,
display high levels of WGA and Con A binding sites at their apical
membranes, whereas the apically-facing membranes of the basal (B)
cells lack these sites (lOVS 31, 294). To determine whether this differ-
ence could reflect differences in glycoprotein complements between the
SB and B cells, the following protocol was carried out. Corneas were
incubated in Jocklick's modified MEM with 5 jiM Ca*^ for 20 hours,
eliciting the release of all SB cells from the corneal surfaces. These cells
and the basal monolayer were then collected and processed in parallel
for Western blot analyses using agglutinins derived from wheat germ
(WGA), Sambucus nigra (SNA), Maakia amurensis (MAA) and peanut
lectin agglutinin (PNA). Lectins were selected because of their intense
and sialic acid-inhibitable binding to the corneal surfaces. SNA and
PNA showed no substantial SB/B differences. MAA showed only quan-
titative increases with stratification. WGA uniquely identified a 94 Kd
band present solely in SB cells not recognized by the other three lectins.
This protein was recovered in the particulate fraction of a cell homoge-
nate from which was completely solubilized by Triton X-100. Removal
of sialic acid by sialidase abolished the ability of WGA binding yet
recovered the PNA binding. These results indicated the existence of a
novel stratification-associated membrane glycoprotein.
Supported by EY 07773 and (CCG) EY 01867.

Interferon-Induced Chemotaxis of la* Langerhans Cells into the Cor-
nea. A. Asbeir", M. S. Jan, and S. P. Epstein (1). Dept. of Ophthalmol-
ogy'", Mount Sinai Hospital, New York, NY.

Purpose. Ocular herpes simplex (HSV) infection induces the migration
of Langerhans cells (LC) into the cornea. HSV keratitis is known to

254

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

stimulate interferon (IFN) production within the cornea. IFN's have
been proposed for treating HSV keratitis. The IFN's were evaluated for
chemotactic activity upon LC. Methods. Mouse corneas were injected
with 2000U IFN-a or 500U IFN-7, each with and without 0.004U In-
terleukin (IL)-ip. 250U IL-2 5jig Cyclosporine A iCsA) and then indi-
rectly stained for la antigen after a 2 h incubation. Results. IFN-a
produced a migration of 1 1.8 ±3.1 LC/comea into the BALB/c corneas
and 8.0 ± 1 LC/comea into the C57BL/6 corneas. In the presence of
5^g CsA, IFN-a failed to elicit any significant levels of chemotaxis in
either strain (BALB/c: 1.0 ± 1; C57BL/6: 1.0 ± 1 LC/comea). IFN-7
elicited the migration of only 0.86 ± 1 LC/comea in the BALB/c and
0.50 ± 0.84 in the C57BL/6 strain. Since IFN-7 has been reported as
having some inhibitory properties, IL-ip or IL-2 was injected into the
same corneas which had been injected with IFN-7. Both those corneas
injected with IFN-7 followed by IL-ip, as well as those injected with
IFN-7 followed by IL-2, induced the migration of =27.7 ± 4 and 27.8 ±
1 (IL-ip: 27.8 ± 1; IL-2: 27.9 ± 2) LC/comea in the C57BL/6 mice.
Conclusions. These results suggest that the IFN's may have a role in
LC migration into the comea in diseases such as herpetic keratitis;
IFN-a by itself, while IFN-7 in conjunction with other cytokines which
may be produced as a result of the ocular infection.

Supported in part by NEI :5P30EYO1867 and Research to Prevent
Blindness, Inc.

comeas were stained for LC utilizing a rat anti-mu la antibody in an
indirect immunoperoxidase technique. Results. As previously reported,
the cytokines, without TGF-p, all induce the migration of la* LC into
the comeas of BALB/c and C57BL/6 mice. IL-ip, in the presence of
TGF-p, failed to be chemotactic (BALB/c: 0.25 ± 0.6; C57BL/6: 0.33 ±
0.7), while IL-2, TNF-a and IL-6 in BALB/a mice all not only remained
chemotactic for LC (28.0 ± 4.9 la* LC/comea), but were elevated to
levels of chemotaxis previously only attainable by IL-2 (25.5 ± 9 la*
LC/comea: p » 0.26). TGF-p with IL-6 in C57BL/6 mice (0.58 ± 0.7 la*
LC/comea) abrogated the chemotactic effect of IL-6 alone (12.2 ± 4 la*
LC/comea: p s 0.0001). Conclusions. Findings suggest that, at least in
BALB/c mice, TGF-p may upregulate LC chemotaxis via the inhibition
of biofeedback mechanisms.

Supported by part by NEI 5P30EYO1867 and Research to Prevent
Blindness, Inc.

Nitric Oxide Synthase: Distribution and Biochemical Properties of
the Enzyme in the Bovine Eye. O. Geyer*, S. M. Podos*, T. W. Mit-
tag* t. Depts. of Ophthalmology* and Pharmacologyt, Mount Sinai
School of Medicine, New York, NY.

Low-Cost 24-Bit Color Contrast Screening for Early Glaucoma. R. M.

Fischer'^', J. Hirsch'^', and S. E. Brodie'". Mount Sinai School of Med-
icine'", and Memorial-Sloan Kettering Cancer Center'^', New York,
NY.

Purpose. Recent reductions in the price of 24-bit "True Color" graphics
adapters for PC-compatible computers have enabled low-cost emula-
tion of the midperipheral color-contrast targets recently described for
the early detection of glaucoma by Arden et al [Clin Vision Sci. 2:303-
320, 1988; lOVS 32:2779-2789, 1991). We have developed such a sys-
tem, which can be installed on a Windows-capable PC with VGA
graphics for an incremental cost of $200. A working version of the
system will be demonstrated. Methods. Tritan-axis stimuli were imple-
mented as large "Landolt C's" against desaturated red and green back-
grounds, matched to chips 85 and 42 from the Farnsworth-Munsell
100-Hue test. Tritan-axis contrast was effected by small variations in
the intensity of the blue component of the stimulus color. For low-
contrast tritan-axis targets near threshold, we found no necessity to
correct these stimuli for luminance variation, due to the minimal con-
tribution to the photopic luminance of the blue stimulus component, as
measured with a calibrated photometer. Results. Trials in patients
with confirmed glaucoma indicate mid-peripheral tritan contrast
thresholds two to three times those in normal controls. Some glaucoma
suspects show lesser degrees of reduction in tritan contrast sensitivity.
Conclusions. 24-bit graphics using the standard VGA monitor is po-
tentially valuable as a low-cost screening tool, within the reach of the
clinical practitioner.

Supported by a grant from Research to Prevent Blindness.

Nitric oxide (NO) is a short-lived free radical that mediates vascular
relaxation via endothelial cells, cytotoxic actions of macrophages and
neutrophils, and may be a transmitted substance in neural tissue. Im-
munohistochemical identification of NO synthase (NOS) showed a
strong presence in the retina and choroid (Yamamato et al, lOVS 33,
Abstr #3604, ARVO 1992). We have investigated NOS levels and basic
biochemical characteristics in various ocular tissues of the bovine eye
by measuring ^[H]citmlline formation that accompanies the formation
of NO from arginine. Retinal NOS was >75% associated with the cy-
tosolic fraction, required O2, L-arginine and NADPH, and was inhib-
ited by N'^nitro-L-arginine. The soluble enzyme required both free
Ca^* and Mg^*, and was inactive when either cation was absent.
However retinal NOS activity was supported by 1-10 \xM Mn^* ions
alone. Compared to control baseline activity (with 4 mM Mg^* , 250 \iM
Ca^*, 450 jxM EDTA present) = 100 ± SEM%, retinal NOS activity
was increased by addition of 1 p.m tetrahydrobiopterin ( +82 ± 9.2%),
was unchanged by the addition of FAD (4 ^.M) + FMN (4 fiM), but was
increased up to 300% when all three cofactors were present. Thus,
electron transfer cofactors may be rate-limiting and availability of
biopterin may be a physiological regulator of NOS. Cytoplasmic frac-
tions of other tissues in the eye also showed NOS activity. Relative to
the retinal activity (2-4 p mol/min/mg protein at optimal substrate,
divalent cation and cofactor conditions) = 100%, the NOS specific ac-
tivity was approx. equal in choroid and retina, was 23 ± 1-4% in
ciliary processes and 34 ± 17.7% in trabecular meshwork. Conclusions.
These results show that nitric oxide synthase is widely distributed in
various ocular tissues. NOS enzymes in the eye may have unique bio-
chemical properties and specific physiological functions in several in-
traocular tissues.

Supported in part by a Castroviejo Fellowship award to OG.

Mediation of Corneal Inflammation through the use of TGF-Beta. A.

Bartolomei, P. A. Asbell, S. P. Epstein. Dept. of Ophthalmology, Mount
Sinai Medical Center, New York, NY.

Purpose. The migration of Langerhans cells (LC) has long been sus-
pected as being important in antigen presentation in surface tissues.
Transforming Growth Factor (TGF)-P is a cytokine, which has a vari-
ety of both inhibitory and immunosuppressive properties. Since comeal
epithelium is normally devoid of la* LC, we investigated whether the
effect of TGF-p on LC chemotaxis in the comea is similar to that of
Cyclosporine A (CsA). Methods. 0.004U recombinant (r) murine (mu)
Interleukin (IL)-ip, 250U r mu IL-2, lOOOU r mu IL-6, or lOU ultra-
pure natural human Tumor Necrosis Factor (TNF)-a were intracome-
ally injected into the comeas of BALB/c and C57BL/6 mice both with
and without 41.5ng natural TGF-p. Following a 2 hour incubation, the

Physiological Evidence for a Membrane-Bound Carbonic Anhydrase
in the Corneal Epithelium. Xiao-Ping Shi, Oscar A. Candia, Aldo Za-
mudio, and Per Wistrand*. Mount Sinai School of Medicine, New York,
NY, and Uppsala University, Uppsala, Sweden*.

Based on the inhibition of transepithelial H^''C03/'''C02 fluxes by ceir-
bonic anhydrase (CA) inhibitors, we predicted that the enzyme must be
present in the comeal epithelium (ARVO 1991). This has now been
confirmed by microchemical detection (Exp. Eye Res., 55:637, 1992).
We also found that under 5% CO2 the unidirectional tear (T) to stroma
(S) flux Ji4c Ts was larger than J14C gx (3.56 vs 2.65 p.Eq/hr ■ cm^). This
net flux was CO2 but not HCO3 dependent. There are no reasonable

Vol. 60 No. 3

ABSTRACTS

255

models by which the cytoplasmic CA can provide directionality to the
CO2 movement. However, this unusual CA-dependent CO2 flux could
be understood if the asymmetry was provided by a membrane-bound
enzyme. If CA resides at the basolateral membrane it could add in the
conversion of HCO3 to CO2 due to the low pH environment just outside
the basolateral membrane. We have tested the effects of two mem-
brane-impermeable CA inhibitors, quaternized sulfanilamide (QS) and
benzolamide, on J 14c st and J^cts across the isolated frog cornea. QS
(10^3 M) reduced Ji4c.ts from 2.95 ± 0.10 to 1.95 ± 0.04 and Ji4c,st
from 1.98 ± 0.09 to 1.68 ± 0.07 jiEq/hr ■ cm^ only when added to the
stromal side. Benzolamide (10 M) had effects on J 14CTS and J 14c, st
similar to QS. We also found that benzolamide inhibits Isc across the
isolated rabbit ciliary epithelium. Based on these results we propose
that there is membrane-bound CA at the frog corneal epithelium and
rabbit ciliary epithelium.

Supported by NIH grants EY04810 and EY00160.

eyes (1267.8 ± 132) was statistically identical (p = 0.94) to the control
eyes which had an average burst pressure of 1262 ± 95.6 mmHg. No
leakage was detected at the wound site at any time before the eyes
reached the maximum pressure obtainable. Conclusions. This study
suggests that intrastromal ring insertion does not alter the structural
integrity of the eye and does not weaken the cornea.

Supported in part by Keravision Inc., Research to Prevent Blind-
ness, Inc., NEI EY01867.

Studies on DARP Expression in Bovine and Human Eyes Using PGR
Based Techniques. J. Danias*, H. Kobayashi*, T. Mittaq* t, S. M. Po-
dos*. Departments of Ophthalmology* & Pharmacology +, Mount Sinai
School of Medicine, New York, NY.

Stratal Distribution and Synthesis Induction of a Tight Junction Pro-
tein in Rabbit Corneal Epithelium. Yu Wang and J. Mario Wolosin.
Mount Sinai School of Medicine, New York, NY.

Tight junctions (TJs) are located between the squamous (SQ) superfi-
cial cells (SPC) of the corneal epithelium. We have previously shown
through measurements of transepithelial resistance that a) digitonin-
elicited devitalization of the SPC layer induces rapid, synthesis-inde-
pendent assembly of TJs (1 hr) in the previous sub-SPC layer and b)
when all three SQ layers are simultaneously devitalized, TJ formation
by the exposed wing (W) cells start after a 2 hr delay, proceed through
the next 10-15 hr and requires gene expression and protein synthesis
(J Membr. B.104, 45; lOVS 32 (suppl), 1976), i.e., proteins critical for
this assembly are synthesized as cells approach the surface in the
course of stratal migration. We have now determined the stratal dis-
tribution of ZO-1, a TJ protein, and its changes during this induced
synthesis-dependent TJ formation. Immunohistology showed: a) no de-
tectable ZO-1 in basal (B) cells; b) scarce punctuate focal accumulations
in the W cells; c) numerous foci in the SQ cells as well as diffuse
cytosolic staining; and d) strong staining at the apical TJ locations in
the SPC layers. Immunoblot analysis of separated layers showed a
B:W:SQ [ZO-11 distribution of 0.02:0.16:1. Devitalization of the SQ lay-
ers induced strong ZO-1 synthesis. Its concentration in the remaining
B and W component (20% of the original total) increased 3 to 5-fold
after 8 and 20 hr, respectively. The synthesis concentrated in the outer
W cells were electron microscopy showed time-dependent development
of TJs. Similar responses were attained in the B cells after removal of
all suprabasal cells by a low [Ca^ * 1 incubation method. These results
provide a biochemical correlate to our previous electrophysiological
studies and demonstrate the value of cell exfoliation methods to study
corneal epithelial cell maturation.

NIH EY 07773 and (CCG) EY 01867.

DARPP-32, a phosphoprotein found in CNS and kidney may regulate
protein phosphatase and Na, K-ATPase. DARP-related proteins in bo-
vine ciliary processes (CP), trabecular meshwork (TM), and aqueous
humor (AH) occur in two distinct forms of ~38kDa and —lOOkDa.
Glycosylatin of the 32 kDa core protein gives rise to the 38 kDa form
(Kobayashi, H. and Mittag, T., lOVS, 33: Abstract #2329, ARVO
1992).

Mouse Ltk cells transfected with DARP-32 cDNA (kindly provided
by Dr. Christian Pifl, Cell Biology, Duke University) were studied to
determine whether DARP-100 also arises by post-translational modi-
fication of a precursor. Western blotting of the Ltk cell proteins with
DARP-specific antibodies stained a prominent band near 32kDa but
not at lOOkDa; evidence against DARP-100 arising from the smaller
DARP-32 precursor, as is the case for the 38kDa form. The absence of
transcripts larger than 1.8 kb in repeated Northern blots of CP mRNA
using DARP-32 coding region probes suggested that DARP-32 and
DARP-100 share some protein homologies but derive from different
genes.

DARPP-32 made in CP and secreted into AH could have a signal
role between CP and TM. We investigated this possibility by amplify-
ing randomly primed cDNA from bovine CP and TM using DARP-32
specific primers. At equal levels of simultaneous amplification of the
internal control protein cyclophyilin, the DARP-32 amplification prod-
uct (213bp) was formed from CP cDNA in large excess compared to TM,
indicating little capacity for TM to synthesize DARP-proteins. To con-
firm synthesis of DARP in CP but not TM in the human eye, we used
the novel technique of in-situ PCR amplification of RNA message.
DARP cDNA was amplified after in-situ reverse transcription on for-
malin fixed, paraffin blocked pathology tissue sections of the human
eye. Sense primers were used as control for the PCR. Human CP epi-
thelial cells were positive for DARP mRNA and thus may be a source
of the DARP-32 present in human AH.

Supported by USPHS (NEI) grant EY-07400 and a Senior Inves-
tigator Award from Research to Prevent Blindness, Inc.

Structural Integrity of the Eye after Intrastromal Corneal Ring In-
sertion. L. G. Alcaraz, A. Bartolomei, C. P. Hurley, and P. A. Asbell.
Department of Ophthalmology, Mount Sinai Hospital, New York, NY.

Purpose. Intrastromal corneal ring insertion is a procedure designed
for the surgical correction of ammetropia. Since previous research in
our laboratory has shown that radial keratotomy in cadaver eyes weak-
ens the eyeglobe, we designed a study to evaluate the effect of the
intrastromal corneal ring on the structural integrity of the eye. Meth-
ods. Twelve eye banks were dehydrated by immersion in dextran so-
lution. Intrastromal corneal rings were inserted in 6 eyes by placing
the plastic ring in a channel made at 2/3 peripheral corneal depth; an
additional six eyes were held as controls without surgical manipula-
tion. In both groups, intraocular pressure was increased by pumping
normal saline through the optic nerve until leakage was noted or max-
imum pressure was reached. This was recorded as the burst pressure of
that eye. Results. Mean burst pressure of the intrastromal corneal ring

Terminal Glycosylation in Rabbit Corneal Epithelium. Meiling Chen
and J. Mario Wolosin. Mount Sinai School of Medicine, New York, NY.

Sugar-sequence specific FITC-lectins were applied to triton x-lOO-de-
lipidated cryostat sections of rabbit corneas to analyze the development
of protein terminal glycosylation in the epithelium as a function of
stratal coordinates. Sambuca nigra agglutinin stained the plasma
membrane (PM) of all the cells within the strata, indicating the pres-
ence of, galactose 2-6 sialyl transferase activity, from the basal (B) cell
upwards. Maakia amurensis (MAA), a marker for galactose 2-3 sialyl-
transferase activity, stained faintly the PM of the B and first supra-
basal (FSB) cell layers, and intensively, all layers above the FSB. Pea-
nut lectin agglutinin (PNA), an indicator of galactose 1-3
galactosyltransferase (1-3 GT) activity stained exclusively intracellu-
lar locations in the FBS layer. The granular appearance of the stain
and its base-to-apex distribution suggests Golgi localization. There was
no substantial stain in either the PM or in the other cell layers. Block-
ing new glycosylation for 4 hr in organ culture with tunicamycin did

256

THE MOUNT SINAI JOURNAL OF MEDICINE

May 1993

not diminished or modified this staining pattern. Because PNA binding
is known to be blocked by sialylation of the terminally exposed galac-
tose these results indicate that the initial stratification event is criti-
cally associated with either a marked activation of a 1-3 GT and/or a
decrease/intracellular relocation of STs allowing sustained accumula-
tion and storage of nascent glycoproteins terminating on galactose res-
idues within the FSB cell. The MAA binding pattern suggests that
upon further upwards migration, increases in STs result in capping
of the exposed terminal galactose residues and possibly, transport to
the PM.

Supported by EY 07773 and (CCG) EY 01867.

Two Different Classes of Adenylyl Cyclase in Ciliary Processes. W. B.
Gud*, K. Kobayashi*, and T. W. Mittaq* t. Departments of Ophthal-
mology*, and Pharmacologyt, Mount Sinai School of Medicine, New
York, NY.

Purpose. Adenylyl cyclase (AC) is thought to play a major role in the
regulation of aqueous humor secretion by ciliary processes. Biochemi-
cal as well as genetic evidence suggests multiple types of membrane-
associated adenylyl cyclase. We have previously demonstrated a bicar-
bonate stimulated AC activity in bovine ciliary processes. This
membrane bound HCOs-sensitive AC is not dependent on GTP and is
additive with the Gs-mediated AC response to isoproterenol (ISO) in
ciliary process particulate fractions (T. W. Mittag, et al., lOVS 33:
ARVO abstract 1056, 1992). Methods and Results. The effects of 9-car-
boxyanthracene (9-CA) and fluorenecarboxylic acids (FC) were inves-
tigated on AC. At least two classes of AC were present in the bovine
ciliary process; type a, activated by receptors coupled to Gs-proteins
(e.g., ISO sensitive AC) and type b, activated by HCO3. Both type a and
type b AC interact with forskolin (FSK) because FSK potentiated AC
responses to ISO and also to HCO3. However 9-CA interacted differ-
ently with the two classes of AC. Addition of 9-CA increased basal and
HCO3 stimulated AC in a concentration-dependent manner (EC50 0.85
mM) by up to + 90% and by two-fold respectively. In contrast, addition
of 9-CA inhibited the ISO and the FSK responses of AC noncompeti-

tively (IC50 5 mM and 10 mM respectively). 9-FC was similarly active
but less so active than 9-CA, while other position isomers (1 and 4
FC) were inactive. Conclusions. These findings indicate that AC
enzymes have a binding site for polyaromatic carboxylic acids of
defined structure. This site can modulate catalytic activity causing
inhibition of type a, Gs-activated AC, but potentiating the type B,
Gs-independent AC.

Supported in part by USPHS grants EY-02619 and EY-07400.

Variability of Optic Nerve Head Parameters Using the Heidelberg
Retinal Tomograph. G. Shafranov, R. A. Schumer, J. S. Lustgarten,
and S. M. Podos. Department of Ophthalmology, Mount Sinai School of
Medicine, New York, NY.

Purpose. We used the Heidelberg retinal tomograph (HRT) to image
optic nerve heads (ONHs) of patients with and without glaucoma. W.;
assessed variability of measurement of various extracted topographic
parameters. Methods. HRT is a confocal laser tomographic imaging
system yielding 32 parallel images each arising largely from the focal
plane. A high-resolution topographic surface representation of the
ONH was derived, and quantitative parameters of interest were ex-
tracted. The 32 images from an area centered on the ONH spanned a
depth range of from 1 mm to 3.5 mm. We used 3 different areas of
measurement: 10° x 10° (n = 13), 15° x 15° (n = 19), or 20° x 20° (n
= 5). There were 9 patients with open-angle glaucoma (GL), 10 suspect
for GL (GS), and 18 who were normal (NL). Results. Mean # images/
eye was 4.7 (range: 3-6). Across all groups, coefficient of variation
(standard deviation ^ mean) (COV) was calculated for: excavation area
(EA): 7.54%; maximum excavation depth: 9.96%; mean excavation
depth: 9.98%; disc volume: 12.2%; excavation volume: 12.2%. For pa-
rameter EA, COV depended on area of measurement: 4.95% (100°^),
7.42% (225°^), or 9.97% (400°^) (p < 0.05). For other parameters, COV
was independent of area of measurement. COVs for GL v. GS v. NL
were not different. Conclusions. Variability of the HRT pareimeters
appears acceptable and not to depend on ONH damage.

Book
Reviews

Gastric Peptic Secretion

R. Cheli, A. Perasso, and G.
Testino, editors. Verona: Cortina
International, 1990. 128 pp, ill.
$50.00. ISBN 88-7749-064-0.

This book is a compilation of 14 short review papers
which were originally presented at a course dedicated
to this subject given at the Advanced School of Oncol-
ogy and Biomedical Sciences at Santa Margherita Lig-
ure, Italy. All the authors are based in Italian centers.
This monograph provides a thorough and comprehen-
sive review of most aspects of pepsin and the pepsino-
gens. The bibliography is critically selected and fairly
complete; however, literature references are not more
recent than 1988 with a few exceptions. The only glar-
ing omission is in regard to the pepsinogen immuno-
histochemical studies of Tatematsu and his Nogoya
group.

The clarity of language varies considerably from
author to author and the language at times is difficult.
Introductory paragraphs precede each short chapter,
and these are sometimes repetitious in recounting his-
torical references.

These shortcomings notwithstanding, the overall
value of the book rates high as a reference source and
provides good critical reviews in most areas of pepsin
and pepsinogen physiology in health and disease. The
papers (chapters) are brief and concise. The authors
provide personal viewpoints along with some well-an-
notated reviews of the literature.

Particularly recommended for the reader's consid-
eration are the chapters on normal morphology (Chap-
ter 1); the regulation of pepsin secretion (Chapter 2);
chief cell mass and pepsinogen I in chronic gastritis
(Chapter 11); and the effect of antisecretory drugs
(Chapter 14).

This monograph presumes a significant level of ex-
pertise on the part of the reader. Serious students of
pepsinogen and pepsin physiology will not be disap-
pointed.

J. L. Werther, M.D.

Clinical Professor of Medicine
Mount Sinai School of Medicine
Clinical Dir., Gastric Cancer Research Laboratory
Division of Gastroenterology
The Mount Sinai Hospital

Non-Responders in
Gastroenterology: Causes
and Treatments

G. Dobrilla, K. D. Bardhan, and
A. Steele, editors. Verona: Cortina
International, 1991. 236 pp, ill.
$174.00. ISBN 88-7749-065-9.

The title of this book, Non-Responders in Gastroenter-
ology, immediately stimulated my interest, since we as
physicians are all faced with patients who simply do
not respond to medical treatment. Nonresponse may
occur for many reasons, including the simplest, such as
noncompliance, to the more difficult, such as subtleties
in diagnosis and pathophysiology. The chapters are
written by a distinguished group of international gas-
troenterologists and surgeons from Australia, Den-
mark, England, France, Germany, and Italy.

The book is organized into eight sections: oral cav-
ity, esophagus, stomach and duodenum, small bowel,
large bowel, rectum, liver, and pancreas. Each section
covers one to three topics.

Topics that I found of interest included the chap-
ters on refractory burning mouth syndrome, alimen-
tary allergies resistant to conventional treatments, and
refractory solitary rectal ulcer syndrome, since these
sections were comprehensive and well organized. The
chapter on refractory duodenal ulcer was quite com-
plete and well referenced. I was surprised and some-
what disappointed that the book's treatment of in-
flammatory bowel disease did not include more
information on the immunosuppressives — 6-MP, aza-
thioprine, cyclosporin — since I believe that these
agents play an important role in treating patients who
have refractory inflammatory bowel disease.

I agree with the editors that "the book lays no
claim to being a comprehensive study of refractoriness
in gastroenterology", but I did enjoy reading this text.
I feel that the book would be an asset to postgraduate
students interested in gastrointestinal diseases, inter-
nists, gastroenterologists, and gastrointestinal sur-
geons who are searching for answers to why patients
may not respond to medical treatments.

Barry Jaffin, M.D.
Clinical Assistant
The Mount Sinai Hospital
Clinical Instructor
Mount Sinai School of Medicine

The Mount Sinai Journal of Medicine Vol. 60 No. 3 May 1993

257

Instructions for Authors

THE MOUNT SINAI JOURNAL OF MEDICINE, established in 1934 as The Journal of the Mount Sinai Hospital, is a peer-re-
viewed general medical journal. It is indexed or abstracted in Index Medicus, Current Contents, Science Citation Index, Hospital
Literature Index, International Nursing Index, Excerpta Medico, Chemical Abstracts, Biological Abstracts, and major databases.
Theoretical and basic science articles for the clinician, reviews, reports of clinical or research experience, articles on any subject
affecting the practice of medicine, short notes in any specialty, book reviews, letters, and articles by medical students are
welcome. The Journal is a participant in the Agreement on Uniform Requirements for Manuscripts Submitted to Biomedical
Journals. The Journal aims at responding to authors of manuscripts within 8 weeks of date of submission.

Title Page. On the title page, on separate lines (double-
! spaced), include (1) title of article, phrased as concisely as pos-
sible; (2) name of each author, including first name and de-
gree; (3) name and address of departments and institutions
from which the work originated; (4) acknowledgments of
sources of support; (5) name, place, and date of any confer-
ences at which paper has been delivered; (6) the statement
Address reprint requests to ... , giving full name and ad-
dress, with zip code, of the appropriate author.

Key Words. Supply 3 to 10 key words or phrases at the bottom
of the title page, for use in indexing the article. Use Med-
ical Subject Headings from Index Medicus.

Units of Measurement. Use SI units, giving range or interval
of normal values.

Illustrations. Photographs, line drawings, graphs, and charts
should increase understanding of the text. Line drawings,
graphs, and charts should be professionally prepared. All il-
lustrations should be submitted in triplicate as sharp, high-
contrast glossy prints. Photomicrographs must be 5V8 in.
wide. On the back of each print affix a gum label with the
number of the figure (numbered consecutively in arable nu-
merals as cited in text), title of manuscript, name of senior
author, and the word "top" with an arrow indicating top edge.
Illustrations which are to appear together should be mounted
accordingly as plates and should correspond to each other in
size. Protect the prints by enclosing them in heavy cardboard;
do not use paper clips. Photographs of patients can be pub-
lished only if a copy of the permission is submitted with the
photograph. Eliminate names of patients and hospital
numbers from x-rays.

Legends. Legends should not duplicate the text. Double space.
Number each legend to match the illustration it accompanies.
Illustrations mounted as plates to appear together should be
accompanied by a single legend identifying the separate ele-
ments by position ("upper left," and so on) or by letters. If
letters are used in the legend, a corresponding letter must
appear on the print itself. Identify all abbreviations. Indicate
magnification and stain for photomicrographs.

Tables. Each table should be typed on a separate page, double
spaced. Number tables consecutively as they appear in text,
using arable numerals ("Table 1"). Give each table a brief
title (for example, "Effect of DMSO on Rats") and type it on a
separate line. If abbreviations or symbols are used in the
table, identify them in a footnote.

1. Nadelman RB, Wormser GP. A clinical approach to Lyme
disease. Mt Sinai J Med 1990;57:144-156.

2. International Committee of Medical Journal Editors. Uni-
form requirements for manuscripts submitted to biomedical
journals. N Engl J Med 1991;324:424-428.

When a manuscript is accepted for publication, you will be
asked to confirm the accuracy of all references.

Proofs. Proofs are sent to the corresponding author fi-om the
printer. Stylistic changes which do not alter meaning should
not be made at this stage. Because of increased printing
charges, the cost of excessive author's alterations beyond rou-
tine correction of typographical errors and major errors in
data may be billed to the author. Proofs should be corrected
and returned to the editorial office within 48 hours.

Reprint Orders. A reprint order form is sent to the corre-
sponding author with proofs; return it to the editorial office.

THE

MOUNT SINAI JO
OF MEDICINE

LEVY LIBRARY. MT SINAI MEDICAL CENTER

3 4805 0315271 3

Forthcoming:

Grand Rounds

Gene Therapy for Immunodeficiency and Cancer

R. Michael Blaese
National Institutes of Health

Neonatal Herpes Simplex Virus Infections:
Natural History, Pathogenesis, and Preventability

Richard J. Whitley

University of Alabama School of Medicine

Stroke Prevention in Atrial Fibrillation

Jonathan L. Halperin
Mount Sinai Medical Center

Theme Issues

SEPTEMBER 1993

GUiality Assurance

Editor: Suzanne Kramer

OCTOBER 1993

Cervical Disk Disease:
Controversies in Neurosurgery

Editors: Kalmon Post and Martin H. Savitz

The Function of the p53 Tumor Suppressor
Gene and Its Clinical Correlates

Arthur Jay Levine
Princeton University

New Developments in the Prevention and
Treatment of Ischemic Stroke

Stanley Tuhrim

Mount Sinai Medical Center

Pathway for Water Absorption in
Isosmotic Transporting Epithelia

Guillermo Whittembury

Instituto Venezolano de Investigaciones Cientificas

NOVEMBER 1993

Update on Treatment of the Elderly
On the 10th Anniversaiy of the
Department of Geriatrics

Editors: Myron Miller and Rein Tideiksaar

JANUARY 1994

Transplantation

Editor: Lewis Burrows

Available now:

Toward the Conquest of Pain

Allan P. Reed, editor
volume 58, no. 3, May 1991
84 pages »15

Festschrift for Rosalyn S. Yalow:
Hormones, Metabolism, and Society

Eugene W. Straus, editor
volume 59, no. 2, March 1992
96 pages S15

Issues in
Medical Ethics

Daniel A, Moros and Rosamond Rhodes, editors
volume 60. no. 1, January 1993
84 pages $15

Subscriptions for 1993: $65 indioidiuils, US.; $70 alt libraries: $75 individuals outside US.

To order subscriptions or copies of back issues ($15 each), please send check, payable to The Mount Sinai Journal of Medicine, to the Journal at Box 1094, One Gustave L Leoy P'jr
new York, NY 10029-6574; counter sales at Suite IE, 50 E. 98th Street, New York, NY (phone: 212 241-6108; FAX: 212 722-6386).

^OUNT SINAI JOURNAL
)F MEDIQNE

OLCIME 60 NUMBER 4

SEPTEMBER 1993

CONTENTS

Contemporary Protozoal Pathogens

Edward J. Bottone, editor

Introduction Edward J. Bottone 259

Free-Living Amebas of the Genera Acanthamoeba and Naegleria:
An Overview and Basic Microbiologic Correlates Edward J.
Bottone 260

Free-Living Amebas: Infection of the Central Nervous Sys-
tem A. Julio Martinez 271

Acanthamoeba Keratitis: There and Back Again Penny A. Asbell 279

Epidemiology of Infections with Free-Living Amebas and Labo-
ratory Diagnosis of Microsporidiosis Govinda S. Visvesvara 283

GRAND ROUNDS

Stroke Prevention in Atrial Fibrillation Jonathan L. Halperin and

Elizabeth B. Rothlauf 289

Frontiers in the Treatment of Ischemic Stroke Stanley Tuhrim 295

GENERAL ARTICLES

Urologic Aspects of Prevention of End-Stage Renal Disease in

Spinal Cord Injury Jose R. Sotolongo, Jr 299

Pathophysiology of End-Stage Renal Disease in Spinal Cord In-
jury N. D. Vaziri 302

Obesity Surgery: Dietary and Psychosocial Expectations and Re-
ality J. Gabrielle Rabner, Sharron Dalton, and Robert J.
Greenstein 305

continued inside

Beth

Veterans Israel
Affairs MEDCAL

¥T^ The Mount Sinai Journal of Medicine is published by The Mount Sinai Medical Center of
I r1 '^^^ "^^^^ following affiliates: Beth Israel Medical Center, New York; Bronx

MOUNT SINAI
JOURNAL OF
MEDICINE

Veterans Affairs Medical Center, New York; and Elmhurst
Hospital Center, New York.

Editor-in-Chief

Sherman Kupfer, M.D.

Editor Emeritus

Lester R. Tuchman, M.D.

Associate Editors

Harriet S. Gilbert, M.D. Julius Wolf, M.D.

Managing Editor

Claire Sotnick

Business and Production Assistant

Karen Schwartz

Assistant Editors

Stephen G. Baum, M.D.
David H. Bechhofer, Ph.D.
Constanin A. Bona, M.D., Ph.D.
Edward J. Bottone, Ph.D.
Jurgen Brosius, Ph.D.
Lewis Burrows. M.D.
Joseph S. Eisenman, Ph.D.
Adrienne M. Fleckman, M.D.
Richard A. Frieden. M.D.
Steven Fruchtman, M.D.
Paul L. Gilbert, M.D.
James H. Godboid, Ph.D.

Richard S. Haber, M.D.
Noam Harpaz, M.D.
Dennis P. Heaiy, Ph.D.
Tomas Heimann, M.D.
Barry W. Jaffin, M.D.
Andrew S. Kaplan, D.D.S.
Samuel Kenan, M.D.
Suzanne Carter Kramer, M.Sc.
Mark G. Lebwohl, M.D.
Kenneth Lieberman, M.D.
Charles Lockwood, M.D.

Lynda R. Mandell, M.D., Ph.D.

Steven Markowitz, M.D.

Bernard Mehl, D.P.S.

Myron Miller, M.D.

Edward Raab, M.D.

Allan Reed, M.D.

Allan E. Rubenstein, M.D.

David B. Sachar, M.D.

Henry Sacks, M.D.

Robert Safirstein, M.D.

Ira Sanders, M.D.

Martin H. Savitz, M.D.
Clyde B. Schechter, M.D.
Michael Serby, M.D.
Phyllis Shaw, Ph.D.
George Silvay. M.D.
Barry D. Stimmel, M.D.
Nelson Stone, M.D.
Max Sung, M.D.
Carl Teplitz, M.D.
Rein Tideiksaar, Ph.D.
Richard P. Wedeen, M.D.

Editorial Board

Barry Freedman, M.B.A.
Richard Gorlin, M.D.
Nathan Kase, M.D.

Panayotis G. Katsoyannis, Ph.D.
Charles K. McSherry. M.D.
Jack G. Rabinowitz, M.D.

John W. Rowe. M.D.
Alan L. Schiller, M.D.
Alan L. Silver, M.D.

Alvin S. Teirstein, M.D.
Rosalyn S. Yalow, Ph.D.

The Mount Sinai Journal of Medicine (ISSN No. 0027-2507; USPS 284-860) is published 6 times a year in January, March, May,
September, October, and November in one indexed volume by the Committee on Medical Education and Publications (Owner), The
Mount Sinai Hospital. Box 1094, One Gustave L. Levy Place, New York, NY 10029-6574. Subscription price: individuals, U.S.,
S65 per year; libraries. $70; individuals outside the U.S., $75. Single copies, $15. New subscriptions begin with the first issue of
the calendar year. Send notice of change of address 60 days before the change is effective. Second-class postage paid at New York,
NY and at additional mailing offices. POSTMASTER: Send address changes to The Mount Sinai Journal of Medicine, Box 1094, One
Gustave L. Levy Place, New York, NY 10029-6574. Copyright 1993 The Mount Sinai Hospital.

Volume 60
Number 4
September 1993

CONTENTS continued

Phase II Trial of Etoposide, Carboplatin, and Ifosfamlde as Sal-
vage Therapy In Advanced Ovarian Carcinoma Ann Marie

Beddoe, Peter R. Dottino, and Carmel J. Cohen 311

The Enterolnsular Axis and Endocrine Pancreatic Function in
Chronic Alcohol Consumers: Evidence for Early Beta-Cell
Hypofunction Renato J. Patto, Ewaldo K. Russo, Durval R.
Borges, and Manoel M. Neves 317

Thoracic Herniated Discs: Review of the Literature and 12
Cases Jeffrey S. Oppenheim, Allen S. Rothman, and Ved P.
Sachdev 321

Screening for Human Immunodeficiency Virus and Sexually
Transmitted Diseases in an Inner-City Colposcopy Clinic

Peter R. Dottino, Rhoda Sperling, and Romina Kee 327

CASE REPORTS

Case Report: Evolution of a Type B Aortic Dissection following
Renal Artery Angioplasty Robert Gendler and Harold A.
Mitty 330

Propofol Use in Malignant Hyperthermia: A Case Report Gor-
don Freedman, Ian H. Sampson, and Steven Cagen 333

THE FORMULARY

Isoniazid-lnduced Hepatotoxicity 337

PRIZES AND AWARDS

Mount Sinai School of Medicine Awards and Prizes Academic

Year 1992/93 338

The Mount Sinai Journal of Medicine Awards 1992 339

IN MEMORIAM

Margit Freund Klemperer Gabriel C. Godman 340

Required Forms for Authors 344

Index to Advertisers

Eli Lilly and Company
Post-Graduate School of Medicine

page 343
page 342

Introduction

Edward J. Bottone, Ph.D.
Director

Clinical Microbiology
Laboratories

The Mount Sinai Hospital
Professor of Microbiology
Professor of Pathology
Mount Sinai School of Medicine

Contemporary

Protozoal

Pathogens

This collection of articles dealing with free-living amebas of the
genera Acanthamoeba and Naegleria, the leptomyxid ameba, and
microsporidia is adapted from presentations at the Workshops in
Clinical Lab Sciences Continuing Education Program, "Contempo-
rary Protozoal Pathogens," held on October 25, 1991 at the Mount
Sinai School of Medicine in New York.

The workshop was indeed special and arose as a consequence
of a paucity of information, among laboratorians and clinicians
alike, about the basic biology, epidemiology, pathogenesis, clinical
presentation, and treatment of free-living ameba infections. The
concept for the program began about a year before this workshop,
when Penny Asbell, an ophthalmologist at Mount Sinai specializ-
ing in corneal infections, saw a patient she suspected of having
Acanthamoeba keratitis and asked the Clinical Microbiology Lab-
oratories to process corneal scrapings in search of this particular
microorganism. The fact was that little microbiologic expertise
could be brought to bear on her request. Furthermore, she sug-
gested cocultivating the corneal scrapings with the bacterium Es-
cherichia coli as a food source. Additionally, she suggested that we
send the patient's contact-lens care system to Govinda S. Visves-
vara at the Parasitic Diseases Branch of the Centers for Disease
Control for isolation of any Acanthamoeba. Not realizing that Vis-
vesvara is a world expert on free-living ameba, we deferred Penny
Asbell's request and I was unsuccessful in recovering Acan-
thamoeba from the corneal scrapings or contact-lens case.

Subsequently, a second patient came to The Mount Sinai Hos-
pital with keratitis presumed to be caused by Acanthamoeba. In
this instance. Dr. M. Nasar Qureshi and I were able to demonstrate
the organism in Giemsa-stained smears of corneal scrapings and in
direct preparations of the patient's contact-lens care system. For us
these findings were stimulating and ushered in a whole new entity
of diagnostic and academic interest. With that in mind, I judged
that if at Mount Sinai there was such a paucity of information on
the basic microbiologic attributes of this microorganism, a discus-
sion of Acanthamoeba might be a good topic to bring to an audience
of colleagues and peers in New York City. Thus, with such naivete,
I approached the codirectors of the continuing education work-
shops, Victor D. Bokkenheuser, Marcelino F. Sierra, and Yvonne
Lue, and we decided to organize this program.

I was assigned the responsibility of securing the various
speakers and promptly called Dr. Visvesvara at the Centers for
Disease Control, who agreed to participate. Because I was inter-
ested in adding discussions of other clinical topics in addition to
keratitis, he suggested I contact A. Julio Martinez, a neuropathol-
ogist at the University of Pittsburgh School of Medicine, who ea-
gerly agreed to participate. I rounded out the program by inviting
Penny Asbell to discuss Acanthamoeba keratitis.

I then embarked on a literature search and was both as-
tounded by the flourishing interest in Acanthamoeba and Naegle-
ria and dazzled by the numerous pioneering contributions, dating
back to 1970, by Visvesvara and Martinez to our understanding of
various laboratory, epidemiologic, and clinical correlates of Acan-
thamoeba and Naegleria infections.

In assembling these discussions I can attest to the joyful ex-
perience M. Nasar Qureshi and I have had in working with Acan-
thamoeba; it is a fascinating entity of significance, especially with
cases of pneumonitis and cerebral involvement being described in
the setting of AIDS.

I hope these discussions will bring a greater appreciation for
these primitive but provocative protozoal species.

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

259

Free-Living Amebas of the Genera
Acanthamoeba and Naegleria:

An Overview and Basic Microbiologic Correlates

Edward J. Bottone, Ph.D.

The major genera containing the pathogenic
free-living amebas are listed in Table 1. Several
species of Acanthamoeba produce human infec-
tion, of which A. astronyxis, A. castellanii, A. cul-
bertsoni, and A. polyphaga are the most common.
Only one species of Naegleria, N. fowleri, has
been associated with human disease, namely, a
fulminating meningoencephalitis (1-3). A^. aus-
traliensis (4, 5) and its subspecies italica (6) have
been shown to produce meningoencephalitis in
animal models that histologically mimics that
produced by Acanthamoeba and N. fowleri in hu-
man hosts. Although neither of these Naegleria
species has been implicated in human disease to
date, there is a potential for these amebas to join
the ranks of N. fowleri with regard to human
pathogenicity. Most recently, Visvesvara and col-
leagues (7) described an ameba of the order Lep-
tomyxida of soil origin which can cause amebic
meningoencephalitis in humans and animals.

The life cycle of free-living amebas is com-
prised of the active trophozoite and a dormant
cyst form; in addition, in Naegleria there is a
transient flagellated form during its life cycle
(Table 2). The trophozoites of both species are
morphologically distinct; they have a nucleus
with a central nucleolus and divide by binary fis-

Adapted from the author's presentation at the Workshops in
Clinical Lab Sciences Continuing Education Program, "Con-
temporary Protozoal Pathogens: Acanthamoeba, Naegleria,
Leptomyxid Ameba, and Microsporidia," at the Mount Sinai
School of Medicine, New York, NY, on October 25, 1991. Final
revision received March 1993. From the Clinical Microbiology
Laboratories, The Mount Sinai Hospital. Address reprint re-
quests to the author at Box 1122, Mount Sinai Medical Center,
One Gustave L. Levy Place, New York, NY, 10029.

sion (Fig, 1), using a mitotic process. The Acan-
thamoeba trophozoite is further distinguished by
the presence of fine filamentous projections called
acanthopodia arising from the cytoplasmic mem-
brane. Cytoplasmic vacuoles may also be promi-
nent. The cyst form resists dessication and allows
the organisms to persist in nature. The cyst is
distinguished by outer (ectocyst) and inner (en-
docyst) walls that communicate by a pore that has
a mucoid plug called an operculum (8). In Acan-
thamoeba the ectocyst wall is polyhedral and the
endocyst is corrugated (Fig. 2), whereas the outer
cyst wall in Naegleria is smooth (7). The factors
that control encystment and excystment are not
known. Generally, adverse environmental condi-
tions induce the trophozoites to encyst, most prob-
ably through a series of complex biochemical
events.

Speciation oi Acanthamoeba is based on cyst-
wall morphology encompassing size and shape
and the distance separating ectocyst and endocyst
wall (9). Because cyst morphology within a single
species may vary according to cultural conditions,
for example monaxenic versus growth media sup-
plemented with MgCl2 (10), other characteristics
have been used for species identification oi Acan-
thamoeba, including isoenzyme profile (11) and
restriction fragment length polymorphism (RLPs)
of mitochondrial DNA (12, 13).

Ecologically, free-living amebas are ubiqui-
tous in the environment. Naegleria species prefer
natural and artificially heated aquatic and soil
environments (3, 14), whereas Acanthamoeba
pervade the entire environment (Table 3). Two
sources from which Acanthamoeba have been iso-
lated, tap water and contact lens cases, bear di-
rectly on the pathogenesis of ameba keratitis.

260

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

Vol. 60 No. 4

PATHOGENIC FREE-LIVING AMEBAS— BOTTONE

261

Further, Acanthamoeba species, as well as other
free-living amebas, have also been isolated from
eyewash stations (15). As a consequence of this
organism's broad distribution in water, sand, soil,
dust, especially in the resistant cyst form, human
contact is constant and probably accounts for the
widespread prevalence of antibody to Acan-
thamoeba and even to Naegleria (16) in human
sera.

Human Infections

The major human infections caused by free-
living amebas of the genera Acanthamoeba and
Naegleria are shown in Table 1. Although A'^.
fowleri produces only one devastating infection,
the rare but highly fatal primary amebic menin-
goencephalitis (PAM), which follows exposure to
amebas in heated water (3, 17), Acanthamoeba
species are more versatile in their human disease

TABLE 1

Free-Living Amebas and Human Infections

Human
infections

Virulence
properties

Acanthamoeba

A. astonyxis
A. castellanii

A. culbertsoni

A. polyphaga

Naegleria

N. fowleri

N. australiensis*
N. australiensis

ssp. italica*
N. gruberi

Leptomyxid ameba

Granulomatous

amebic

encephalitis
Keratitis
Pneumonitis

Granulomatous
dermatitis

Isolated from:
ear,

nasopharynx,

maxillary

sinus,

mandibular

autograft,

stools

Primary amebic
meningo-
encephalitis

Adhere to

mucosal

surfaces
Migrate through

tissue (a 37°C,

28°C
Cytopathic

enzymes

elastase,

protease,

coUagenase

Adhere to

mucosal

surfaces
Migrate through

tissue (aj 37°C
Cytopathic

enzymes

elastase,

protease
Phagocytosis of

host

neutrophils

* Produces in mice subacute meningoencephalitis comparable
to that caused by Acanthamoeba spp.

potential. Indeed, Acanthamoeba can also invade
the central nervous system to produce granulo-
matous amebic encephalitis (GAE), predomi-
nantly in debilitated hosts. In this setting, how-
ever, the clinical manifestations of a subacute,
diffuse, necrotizing, granulomatous encephalitis
differ significantly from the acute naeglerial me-
ningoencephalitis. The route of acquisition ap-
pears to be by hematogenous spread of Acan-
thamoeba from either a pulmonary or cutaneous
focus (18), rather than through the nasal mucosa
and along the olfactory tracts with eventual in-
vasion of the subarachnoid space, as in naeglerial
meningoencephalitis (19-21).

Diagnostically, in PAM, Naegleria trophozo-
ites are abundant in brain tissue and cerebrospi-
nal fluid, whereas in GAE, Acanthamoeba tropho-
zoites and cysts are easily seen in histologic
sections, but Acanthamoeba trophozoites are ab-
sent from cerebrospinal fluid (22). Recently,
Acanthamoeba central nervous system infection
has increased in the setting of the acquired
immunodeficiency syndrome (23-26). Acan-
thamoeba have also caused disseminated infec-
tions involving the skin (27). Other sites of in-
volvement include the sinuses (23, 24),
nasopharynx (28), external ear canal (29), gastro-
intestinal tract (30), and autogenous mandibular
bone graft (31).

Acanthamoeba keratitis is now the most her-
alded of human infections, occurring in an in-
creasing number of individuals predominantly in
association with daily wear of soft contact lenses
(32). Since the first reported case of Acan-
thamoeba keratitis, documented in 1973 (33), nu-
merous case reports have appeared, and the topic
has been extensively reviewed by Auran et al.
(34) and Asbell (35).

The role of Acanthamoeba in keratitis is un-
disputed and is substantiated mainly by human
clinical cases and by animal experimentation
(36). Despite these revelations, however, the ex-
act pathogenesis of human eye infection remains
unresolved. In the nonwearer of contact lenses,
antecedent minor corneal trauma is thought to be
a prerequisite to Acanthamoeba keratitis. Be-
cause most of these infections occur during
warmer weather, often subsequent to water expo-
sure (34), it is theorized, but unproven, that the
amebic trophozoite is the infectious form.

In the contact-lens wearer, the pathogenesis
of amebic keratitis is also unresolved, although
John and colleagues (37) have offered a plausible
sequence of events. In their opinion, on exposure
of the contact lens to solutions containing Acan-
thamoeba, cysts and trophozoites almost immedi-

262

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

TABLE 2

Life Cycle o/"Acanthamoeba*

Trophozoite

Single nucleus; large, dense, central nucleolus

Flat, irregular, produce slender pointed micropseudopodia

(acanthopodia)
Divides rapidly by binary fission (mitosis)

Cyst form

Single nucleus; central nucleolus; polyhedral, convex,

double-walled cyst
Outer, ectocyst
Inner, endocyst

Cyst walls communicate through pores containing a plug,
operculum

* During life cycle of Naegleria, a transient pear-shaped bi-
flagellate stage is produced. Cysts are smooth-walled.

ately adhere to the lens surface. This is true even
for unworn contact lenses, extended-wear lenses,
and disposable soft contact lenses (38). In this set-
ting Acanthamoeba may persist if the lens is not
optimally disinfected. Placement of the lens on
the eye surface allows amebas to become estab-
lished in the conjunctival flora and brings Acan-
thamoeba into close proximity to the corneal
stroma. Minor corneal irritation or abrasion may
allow corneal penetration of the amebas.

The mechanism proposed by John et al. (37)
to account for ameba keratitis gives little consid-
eration to the role of contaminating microflora in
the eye itself or in the contact lens care system as
a major cofactor in the provision of numerous
ameba trophozoites as a prelude to corneal inva-
sion.

At The Mount Sinai Hospital, we encoun-
tered a patient with ameba keratitis whose con-
tact lens care system was cocontaminated with an
Acanthamoeba and numerous bacteria that
proved on culture to be Xanthomonas malto-
philia. Of striking interest on gram-stained
smears of the solution was the presence of numer-
ous Acanthamoeba trophozoites encircled by a
ring of gram-negative slender rods, many inter-
nalized in phagocytic vacuoles (Fig. 3). Investi-
gation of the role of bacterial cocontaminants in
contact lens care systems as substrates for the
growth of Acanthamoeba revealed that Acan-
thamoeba trophozoites readily feed on bacterial
cocontaminants deriving from aquatic and soil
environments cohabitated by amebas (39). These
bacterial species include X. maltophilia, Flavo-
bacterium breve, and Pseudomonas paucimobilis,
which are common contaminants of contact lens
care systems (40, 41). Thus, the nature of the bac-
terial cocontaminants in the contact lens case ei-
ther promotes or abates Acanthamoeba growth.

TABLE 3

Ecology o/" Acanthamoeba

Free-living ameba present in all types of environments

throughout world.
Cyst form resistant to disinfection and dessication.
Isolated from:

Fresh water, seawater, ocean sediment, swimming water.

Tapwater, bottled mineral water.

Industrial cooling water, air conditioners.

Air, sewage, soil, compost, dust.

Chlorinated swimming pools, physiotherapy pools.

Dental treatment units, dialysis units.

Contact lens cases. Eye wash stations

Vegetables, fish, reptiles, birds.

a fundamental principle underscoring Acan-
thamoeba-indxiced keratitis.

In the presence of a suitable bacterial food
source, Acanthamoeba cysts undergo excystment
(42), freeing the predatory trophozoite form,
which then feeds on the bacterial substrate and
multiplies to large numbers (Fig. 4). Attachment
of Acanthamoeba trophozoites and cysts to the
contact lens itself (37, 38) now apposes numerous
amebas contiguous to the corneal surface poised
for invasion subsequent to a minor corneal insult.
The end result of these ameba-bacterial interac-
tions, initiated in the ecologic niche of contact
lens care system, is clinically apparent keratitis.

Virulence

The attributes of free-living amebas contrib-
uting to virulence in human infection are listed in
Table 1. Initially, as with any microbial infection,
adherence of the offending microbe to the target
epithelium is a prerequisite for infection. Acan-
thamoeba have been shown to adhere not only to
inanimate surfaces (37, 38) but also to corneal
epithelium (43), a characteristic which increases
with time (120 min) and incubation temperature
(37°C) (44). Interestingly, Acanthamoeba species
may show interstrain differences in adherence to
corneal epithelium, a correlate which may under-
score the increased incidence of specific Acan-
thamoeba species, especially A. polyphaga and A.
castellanii, in keratitis (44).

Migration patterns of Acanthamoeba and
Naegleria through tissue may also account for dif-
ferences in clinical presentation of disease and for
the epidemiologic circumstance surrounding ac-
quisition of a pathogenic free-living ameba. N.
fowleri, the causative agent of PAM, a rapidly
fatal meningoencephalitis, is often acquired by
swimming in contaminated warm waters. Infec-
tion occurs when the nasal mucosa is exposed to
the trophozoite form of this ameba, which then

Vol. 60 No. 4

PATHOGENIC FREE-LIVING AMEBAS— BOTTONE

263

penetrates the nasopharyngeal mucosa, migrates
along the olfactory nerves, and invades the brain
through the cribiform plate. At 37°C, N. fowleri
trophozoites exhibit rapid locomotion, which aids
their invasive potential (45), whereas nonpatho-
genic Naegleria species, such as N. gruberii, mi-
grate poorly at 37°C but are more active at 28°C.

Consequently, N. fowleri produces one major
human disease, whereas Acani/iamoe6a induces a
broader clinical spectrum, including cutaneous
lesions (27). Furthermore, the epidemiology of
Acanthamoeba brain involvement, granuloma-
tous amebic meningoencephalitis, is different
from that of Naegleria. Indeed, Acanthamoeba ce-
rebral infection occurs in the absence of exposure
to contaminated water by hematogenous spread
from a cutaneous or pulmonary focus (46). Per-
haps the cutaneous propensity reflects the predi-
lection for Acanthamoeba to migrate better at
lower (28°C) than at higher (37°C) temperatures.

Penetration of mucosal surfaces subsequent
to adherence and colonization is an essential step
in the invasive process of free-living amebas. To
achieve tissue invasion, Naegleria and Acan-

thamoeba species have been shown to elaborate
several cytopathic enzymes, such as phospholi-
pase A, sphingomyelinase, protease, and an
elastase (47). Analogous to human leukocyte
elastase, the amebic enzyme degrades connective
tissue proteins such as fibrinogen, collagen, and
proteoglycan. A. castellanii has additionally been
shown to produce a collagenolytic enzyme that
digests collagen shields and purified collagen and
may contribute to corneal stromal degradation (48).

Phagocytosis of human neutrophils (49) and
erythrocytes (50) through contact-dependent phe-
nomena (analogous to the behavior of Entamoeba
histolytica) may also underscore the pathogenic-
ity, invasiveness, and virulence of free-living
amebas. Phagocytosis of leukocytes and erythro-
cytes as well as cultured mammalian tissue (51)
may be mediated through suckerlike structures
(amoebostomes) observed by electron microscopy
on the surface of N. fowleri (52).

Host Immune Responses

Despite the prevalence of free-living amebas
in the human environment, few clinical cases

Fig. 1. Left Phase microscopy photomicrograph
showing dividing Acanthamoeba linked by remnant
of cytoplasmic membrane. Fig. 2. Left below
Electron photomicrograph of Acanthamoeba cysts
showing characteristic double-walled structure.
Outer wall (ectocyst) is hexagonal, inner wall (en-
docyst) circumscribes the contour of ameba tro-
phozoite (original magnification x5200). Fig. 3.
Above Gram-stained smear of contact lens care so-
lution contaminated with Acanthamoeba and bacte-
ria showing mantle of adherent gram-negative ba-
cilli and some within phagocytic vacuoles in
cytoplasm.

264

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

Swimming in pond Removal of lens

Trophozoite attachment Acanthamoeba
to contact lens keratitis

Fig. 4. Role of bacterial cocontaminants in contact lens care
systems leading to Acanthamoeba keratitis. Appropriate bac-
terial cocontaminants, usually from aqueous source, support
growth oi Acanthamoeba, leading to increased inoculum den-
sities of cysts and trophozoites which can adhere to contact
lens. Placement of lens over corneal abrasion facilitates
ameba penetration of cornea, resulting in keratitis. Repro-
duced with permission from J Clin Microbiol 1992; 30:2447-
2450. Original sketch courtesy of Robert M. Madayag.

have been described. Recorded clinical cases have
occurred in the setting of immunosuppression
(Acanthamoeba) (28), in immunologically de-
prived sites such as the cornea, or sporadically in
previously healthy individuals, for example Nae-
gleria meningoencephalitis. To account for the
sparsity of free-living amebic human infections,
several host defense mechanisms have been found
operative (Table 4).

TABLE 4

Host Defense Against Acanthamoeba and
Naegleria Infections

Serum antibodies

IgG, IgM; higher against Naegleria; elicited in response to

environmental exposure to free-living amebas
Inhibit adherence of amebas to human tissue
Opsonizing

Complement activation

Alternative pathway

In vitro lysis of amebas in presence of C5-C9
Generation of iC3b on ameba membrane for recognition by
phagocytes

Highly pathogenic Naegleria fowleri resistant to
complement-mediated lysis even in presence of specific
antibodies

Phagocytic cells (neutrophils)

Cytokine (TNF-a); altered neutrophils in presence of either

antibody or complement
Neutrophils surround and adhere to ameba surface through
pseudopod extension
Activation of:

Oxidative respiratory system 02~ • H2O2
Myeloperoxidase, H2O2, halide system

Complement-Mediated Lysis. Initially, in
the nonimmune host, Acanthamoeba species may
activate the alternative complement pathway,
which leads to lysis of amebas (53). From a con-
ceptual point of view it is interesting to speculate
that this primitive defense mechanism may have
evolved concomitant with, or in response to, the
equally primordial one-celled ameba.

Although complement activation by Acan-
thamoeba may ensue through the classic anti-
body-mediated or the alternative pathway, there
is no clear-cut evidence to support complement-
mediated in vivo lysis of amebas (54). Indeed, ac-
cording to Ferrante (54), it appears that the main
function of complement activation is to generate
opsonic factor iC3b to enhance phagocytic cell rec-
ognition of and attachment to amebas.

In contrast to Acanthamoeba and weakly
pathogenic Naegleria species, highly pathogenic
N. fowleri amebas are resistant to in vitro com-
plement-mediated lysis (55). Thus, although
pathogenic N. fowleri do activate complement,
they may evade lysis by inhibiting the terminal
steps of complement action, by shedding the
membrane attack complex, or by internalizing
and degrading the terminal complement compo-
nents (55).

Serum Antibodies. Because of the ubiquity of
free-living amebas, antibodies against these spe-
cies are prevalent in human sera. In a study con-
ducted in New Zealand (16), Cursons et al. found
antibodies in all 93 serum samples tested with

Vol. 60 No. 4

PATHOGENIC FREE-LIVING AMEBAS— BOTTONE

265

titers ranging from 1:5 to 1:20 for Naegleria spe-
cies and from 1:20 to 1:80 for Acanthamoeba spe-
cies. Antibodies belonged primarily to the immu-
noglobulin G (IgG) and immunoglobulin M (IgM)
classes, suggesting both old and recent contact.
Further, IgG antibodies were detected in cord
sera of newborn infants, suggesting that antibod-
ies to Acanthamoeba are transferred placentally.
Cursons et al. (16) postulate that serum antibod-
ies to Acanthamoeba may serve as opsonins en-
hancing phagocytosis of amebas, may prevent at-
tachment of amebas to mucosal surfaces, can
activate complement, and can agglutinate or im-
mobilize amebas.

Serum antibodies have also been detected
against N. fowleri and the nonpathogenic N. gru-
beri. Unlike antibodies to Acanthamoeba, agglu-
tinating antibodies are negligible in infants' sera
because these antibodies are of the IgM class and
hence not passed transplacentally (56). Animal
studies have shown that resistance to intranasal
challenge by N. fowleri could be enhanced with
immune sera, thereby supporting a role for anti-
body in immunity to this ameba species (57). In-
terestingly, A^. fowleri may evade immune sur-
veillance by removing antibody from its surface
by redistributing, capping, and internalizing sur-
face-bound antibody (58).

Neutrophils. In host immunity to free-living
ameba infections, the neutrophil plays a central
role. Early and elegant work of Ferrante and
Thong (59) demonstrated that in immune mice,
neutrophils surround, adhere to, and immobilize
ameba trophozoites, and pinch off and ingest por-
tions of the trophozoite, which then ruptures. Fur-
ther, in support of the central role of the neutro-
phil, Ferrante and colleagues (60) showed that
either abrogation of neutrophil function or deple-
tion of neutrophils in immune mice markedly re-
duced immunity to pathogenic Acanthamoeba
and Naegleria. Macrophages and lymphocytes
from human peripheral blood are not in them-
selves effector cells against free-living amebas
even in the presence of antibody (61, 62).

For human neutrophils to kill Acanthamoeba
and Naegleria effectively, several other host de-
fense parameters must be met. Initially, neutro-
phils must bind to the ameba trophozoite surface,
a feat achieved in the presence of antibody or
complement components mediating neutrophil
attachment. Second, neutrophils must be acti-
vated by the cytokine tumor necrosis factor alpha
(TNF-a), derived from stimulated macrophages
(63).

In studies with N. fowleri, Ferrante and col-

TABLE 5

Diagnosis of Free-Living Ameba Infections

Direct staining of corneal scrapings/CSP^
Giemsa

Cysts clear, refractile, polyhedral, stellate.

Trophozoite pale blue, uninucleate, may resemble
macrophages
Gram

Cysts poorly staining

Trophozoites poorly staining
Fluorescent antibody (indirect)

Calcofluor white (fluorescent fabric brightener with affinity
for chitin and cellulose [B-linked carbohydrates])
Cysts, corrugated double wall, apple-green

fluorescence; interior, orange-red
Trophozoites counterstained by Evans blue a bright
red-orange

Fluorescein-conjugated lectins

Glycoproteins that bind to carbohydrate
moieties — Concanavalin A; cysts and trophozoites stained

Fixed-tissue stains
Hematoxylin-eosin
Periodic acid-Schiff
Gomori's methamine silver

Examination of contact lens

by light microscopy or by calcofluor staining; cysts easily
visible

Examination of contact lens care system
Cultural methods

Nonnutrient agar plates with Page's ameba saline seeded
with Escherichia coli or Enterohacter aerogenes (18—24 h
old)

Scan agar surface for trophozoites and cysts (3 days
incubation) for 7 days; trophozoites produce wave-like
tracks
Filter-culture technique
Specimen collected in ameba saline; filtered, 0.22 ftm
pore size; filter removed and placed on E. coli seeded
plates

Amplification of ameba by direct addition of
growth-promoting bacterial species to specimen/contact
care system solution

Tissue culture, monkey kidney (vero)
cytopathic effect, plaques

Animal inoculation

Two-week-old mice, nasal inoculation with ameba
suspension

leagues (64) showed that primed neutrophils kill
amebas through enhancement of hypochlorite
(HOCl) production through the myeloperoxidase
H202-halide system, and enhancement of the ox-
idative respiratory system leading to increased
superoxide anion production. In the presence of
amebas opsonized by either antibody or comple-
ment, similar augmentation of neutrophil amebi-
cidal activity by TNF-a has been demonstrated
against Acanthamoeba (A. culbertsoni) as well
(54).

266 THE MOUNT SINAI JOURNAL OF MEDICINE

Fig. 5. Giemsa-stained smear of corneal scraping showing
Acanthamoeba trophozoite distinguished by central dense nu-
cleolus, numerous vacuoles, and ovoid ameba morphology.

Diagnosis

The diagnosis of Acanthamoeba or Naegleria
infection requires an initial heightened index of
suspicion for their presence. Whereas Naegleria
may be suspected on epidemiologic evidence — for
example meningoencephalitis following exposure
to amebas in w^arm contaminated water —
nonkeratitic Acanthamoeba infections require an
awareness of the potential presence of Acan-
thamoeba.

Acanthamoeba keratitis is the most com-
monly recognized presentation of this free-living
ameba. Diagnosis ensues from an initial index of
suspicion through a variety of staining, histo-
logic, and cultural modalities (64a) (Table 5). An-
imal inoculation is rarely utilized.

Standard Staining Techniques. When Acan-
thamoeba keratitis is suspected, the ophthalmol-
ogist should alert the microbiologist and should
forward to the laboratory corneal scrapings and
the patient's contact lenses and contact lens stor-
age system to be microbiologically evaluated for
Acanthamoeba (65). At the outset, corneal scrap-
ings may be examined after staining by the Gram
and Giemsa methods. The former is of value in
ruling out a bacterial or fungal cause of the cor-
neal ulcer; the latter may be used to detect mul-
tinucleated giant cells and intranuclear inclu-
sion, suggesting a viral (herpes, adenovirus)
keratitis.

In gram-stained preparations Acanthamoe-
ba trophozoites may be only faintly visible, al-

September 1993

though an uneven cellular contour due to pseudo-
pod extension and a nucleus with a central nucle-
olus are readily visible. Further, food vacuoles
may be discerned in the trophozoite cytoplasm. In
our laboratory, basic fuchsin is used as the gram
counterstain, and by overstaining the corneal
smear for two minutes, ameba trophozoites are
more readily discerned. Careful search should
also be made for the copresence oi Acanthamoeba
cysts delineated by their double-walled stellate
morphology.

In Giemsa-stained corneal scrapings, Acan-
thamoeba trophozoites are pale lavender, irregu-
lar, and surrounded by a clear halo (Fig. 5). Al-
though difficulty may be encountered in
differentiating trophozoites from mononuclear
and epithelial cells, careful attention to nuclear
morphology (irregular and kidney-shaped in
monocytes), the presence of large food vacuoles,
and extending pseudopodia may aid Acan-
thamoeba differentiation. In this staining tech-
nique Acanthamoeba cysts are clear, refractile,
and either faintly stained or unstained (66).
Other staining techniques used with variable
success in detecting Acanthamoeba include he-
matoxylin-eosin, periodic acid-Schiff, Gomori's
methamine silver, Wright's stain, methylene
blue, Congo red, Janus green, Masson stain, and
Wheatley's trichromes, and acridine orange.

Newer Staining Techniques. Because of the
inherent expertise required for the recognition of
Acanthamoeba directly in corneal scrapings
through evaluation of standard staining tech-
niques, several newer techniques that enhance
the visualization of Acanthamoeba trophozoites
and cysts have been described. One makes use of
calcofluor white (67), a fluorescent brightener
with a strong affinity for chitin and cellulose,
polymers of B-linked polysaccharides. Initially
used to stain cell walls of fungi containing these
compounds, calcofluor white stains the amebic
cyst wall a fluorescent green while the proto-
plasm stains red-orange (67). Evans blue is used
to counterstain background epithelial and in-
flammatory cells as well as the ameba trophozoite
a bright red-orange (67, 68). Because of the affin-
ity of calcofluor white for chitin, cysts are more
readily detected than the amebic form (69). Cal-
cofluor white alone without counterstain has also
revealed Acanthamoeba cysts adherent to the sur-
face of a contact lens derived from a patient with
suspected Acanthamoeba keratitis, thereby con-
firming the diagnosis (70).

Indirect fluorescent antibody staining of
Acanthamoeba using species-specific hyperim-
mune antisera has been used mainly to identify

Vol. 60 No. 4

PATHOGENIC FREE-LIVING AMEBAS— BOTTONE

267

Acanthamoeba isolated from human infections
(71). This technique has also allowed for the de-
tection of Acanthamoeba trophozoites and cysts in
various human and experimentally infected tis-
sues (71, 72). A limitation of this technique is the
unavailability of commercial antisera.

In an attempt to remedy the lack of a widely
available technique to identify both forms of
Acanthamoeba, Robin and colleagues (73) inves-
tigated a variety of fluorescein-conjugated lec-
tins, such as concanavalin A and wheat germ ag-
glutinin. Lectins are glycoproteins which bind
specifically to carbohydrate moieties. Conjugat-
ing these compounds to a fluorescent dye facili-
tates their visualization when bound to the sur-
face of cells (bacteria, fungi, ameba) containing
appropriate lectin receptors. Each of the nontra-
ditional staining methods have proven successful
in demonstrating Acanthamoeba in deparaf-
finized histologic sections of corneal or other tis-
sue biopsies.

Cultural Assays. Culturally, Acanthamoeba
may grow alone on blood or chocolate agar, pro-
ducing tracks across the agar surface (65). This
cultural technique, however, is quite insensitive
and in the absence of a clear suspicion of the pres-
ence of Acanthamoeba in the inoculated speci-
men, these tracks may be easily overlooked on
plates discarded without microscopic examina-
tion of the agar surface.

To enhance cultivation of Acanthamoeba
from corneal scrapings, it is recommended that
the specimen be inoculated to a nonnutrient agar
preseeded with either Escherichia coli or Entero-
bacter aerogenes (74). In this fashion an appropri-
ate bacterial lawn is made available for the pred-
atory ameba trophozoite to feed and grow on.
Agar plates are then scanned by low-power
(100 X) microscope in a search for ameba tropho-
zoites. Because Acanthamoeba are free-living,
cocultivation with bacterial species derived from
their natural environment, such as Xanthomonas
maltophilia, Pseudomonas paucimobilis (but not
P. aeruginosa) (74a), and Flauobacterium breve,
which are also common contaminants of contact
lens care systems (39, 40), enhances growth of
Acanthamoeba species (39). Gradus et al. (75)
have emphasized the value of filtering corneal
scrapings inoculated to Page's ameba saline (74),
placing the filter on a lawn of E. coli, and exam-
ining the medium daily by low-power microscopy
for trophozoites.

Recently, Bottone and colleagues (76) de-
scribed an amplification technique for the culti-
vation of Acanthamoeba from corneal scrapings
and contact lens care systems. In this procedure.

Fig. 6. Phase-contrast photomicrograph showing numerous
Acanthamoeba cysts and trophozoites following amplification
of growth in presence of Xanthomonas maltophilia bacterial
food source.

specimens are inoculated to tubes of sterile phys-
iologic saline (2.5 mL), after which 0.5 mL of a
fresh turbid suspension of X. maltophilia is added
and incubated at 37°C. A drop of the cocultivation
suspension is then examined at daily intervals by
direct or phase contrast microscopy for the pres-
ence of, or increased number of, amebic trophozo-
ites and cysts (Fig. 6). The advantage of this pro-
cedure over the agar method is that it obviates
the low-power scanning of agar surfaces in search
of ameba trophozoites which, in a liquid milieu,
are more readily discerned as contrasted to the
more restrictive agar surface. Whatever the cul-
tural technique, the sensitivity of the assay is di-
rectly related to the clinical suspicion of the pres-
ence of Acanthamoeba (35).

In vitro tissue culture techniques using hu-
man corneal epithelial cells (77) or rabbit corneal
fibroblasts (78) have also been described for the
isolation of acanthamoebae but remain limited to
research rather than routine use in laboratories.

Summary and Conclusions

Free-living amebas of the genera Acan-
thamoeba and Naegleria are allied in basic cell
biology, ecology, and human disease-producing
potential. However, several enigmas surrounding
these amebas need further intellectual and scien-
tific scrutiny. For instance, a clearer differentia-
tion is needed of factors delineating pathogenic
species — those most frequently associated with
human infections — from nonpathogenic species.
Further, have the pathogenic species bridged the
gap in a step-wise fashion between their habitat
and the human host to express their disease-pro-
ducing potential?

Second, what attributes of amebas account
for the spectrum of disease caused by Acan-

268

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

thamoeba and Naegleria? In the brain, Naegleria
is highly destructive of tissue in the course of a
rapidly evolving, hemorrhagic primary meningo-
encephalitis in normal individuals. Central ner-
vous system invasion by Acanthamoeba, however,
occurs only in compromised hosts, and is a more
slowly evolving subacute to chronic encephalitis.
Further, the epidemiology of the infections are
disparate. Naegleria is acquired from exposure to
contaminated fresh water; this epidemiologic link
is absent in Acanthamoeba meningoencephalitis.
Naegleria invades the central nervous system via
the olfactory nerve; Acanthamoeba is deposited in
the central nervous system via hematogenous
spread from a pulmonary or cutaneous focus. Ad-
ditionally, Naegleria is apparently restricted to
the brain; Acanthamoeba is not so bridled, and
invades more sites in the human body.

Finally, Acanthamoeba are beginning to ap-
pear opportunistically more frequently in pa-
tients infected with the human immunodeficiency
virus. To avoid the same lack of effective thera-
peutics we still face in treating Cryptosporidium
infections in these patients, more emphasis must
be placed on clinical trials dealing with the man-
agement of Acanthamoeba infections in patients
with AIDS.

References

1. Fowler M, Carter RF. Acute pyogenic meningitis probably

due to Acanthamoeba. Brit Med J 1965; 2:740-742.

2. Robert VB, Rorke LB. Primary amebic encephalitis, prob-

ably from Acanthamoeba. Ann Int Med 1973; 79:176-
179.

3. John DT. Primary amebic meningoencephalitis and the

biology of Naegleria fowleri. Annu Rev Microbiol 1982;
36:101-123.

4. De Jonckheere JR. Naegleria australiensis sp. nov., an-

other pathogenic Naegleria from water. Protistologica
1981; 17:423-429.

5. De Jonckheere JF, Aerts M, Martinez AJ. Naegleria aus-

traliensis: experimental meningoencephalitis in mice.
Trans R Soc Trop Med Hyg 1983; 77:712-716.

6. Scaglia M, Gotti S, Cevini C, Bernuzzi AM, Martinez AJ.

Naegleria australiensis ssp. italica: experimental study
in mice. Exp Parasitol 1989; 69:294-299.

7. Visvesvara OS, Martinez AJ, Schuster FL, Leitch GJ,

Wallace SV, Sawyer TK, Anderson M. Leptomyxid
ameba, a new agent of amebic meningoencephalitis in
humans and animals. J Clin Microbiol 1990; 28:2750-
2755.

8. Visvesvara GS. Classification of Acanthamoeba. Rev In-

fect Dis 1991; 13:8369-372.

9. Pussard M, Pons R. Morphologic de la parol cystique et

taxonomie du genre Acanthamoeba (Protozoa, Amoe-
bida). Protistologica 1977; 13:557-598.

10. Stratford MP, Griffiths AJ. Variations in the properties

and morphology of cysts of Acanthameba castellanii. J
Gen Microbiol 1978; 108:33-37.

11. Costas M, Griffiths AJ. The suitability of starch-gel elec-

trophoresis of esterases and acid phosphatases for the

study of Acanthamoeba taxonomy. Arch Prostisterk
1980; 123:2727-2729.

12. Bolger SA, Zarley CD, Burianek LL, Fuerst PA, Byers TJ.

Interstrain mitochondrial DNA polymorphism detected
in Acanthamoeba by restriction endonuclease analysis.
Mol Biochem Parasitol 1982; 8:145-163.

13. McLaughlin GL, Brandt FH, Visvesvara GS. Restriction

fragment length polymorphisms of the DNA of selected
Naegleria and Acanthamoeba amebae. J Clin Microbiol
1988; 26:1655-1658.

14. Hurzianga HW, McLaughlin GL. Thermal ecology of Nae-

gleria fowleri from a power plant cooling reservoir. Appl
Environ Microbiol 1990; 56:2200-2205.

15. Paszilo-Kolua C, Yamamoto H, Shahamat M, Sawyer TK,

Morris G, Colwell RR. Isolation of amoebae and
Pseudomonas and Legionella spp. from eyewash sta-
tions. Appl Environ Microbiol 1991; 57:163-167.

16. Cursons RTM, Brown TJ, Keys EA, Moriarty KM, Till D.

Immunity to pathogenic free-living amoebae: role of hu-
moral antibody. Infect Immun 1980; 29:401^07.

17. Martinez AJ. Free-living amebas: natural history, pre-

vention, diagnosis, pathology and treatment of disease.
Boca Raton, Florida: CRC Press, 1985.

18. Robert VB, Rorke LB. Primary amebic encephalitis, prob-

ably from Acanthamaoeba. Ann Intern Med 1973; 79:
174-179.

19. Martinez AJ, Nelson EC, Jones MM, Duma R, Rosenblum

J. Experimental Naegleria meningoencephalitis in
mice. An electron microscope study. Lab Invest 1971;
25:465^75.

20. Martinez AJ, Duma RJ, Nelson EC, Morretta FL. Exper-

imental Naegleria meningoencephalitis in mice. Pene-
tration of the olfactory mucosal epithelium by Naegle-
ria and the pathologic changes produced: a light and
electron microscope study. Lab Invest 1973; 29:121-
133.

21. Sarphie TG, Allen DJ. Scanning electron microscopy of

Acanthamoeba culbertsoni as observed in the subarach-
noid space. Am J Clin Pathol 1977; 68:485--492.

22. Jones DB. Acanthamoeba — the ultimate opportunist. Am

J Ophthalmol 1986; 102:527-530.

23. Gonzalez MM, Goold E, Dickinson G, Martinez AJ, Vis-

vesvara G, Cleary TJ, Hensley GT. Acquired immuno-
deficiency syndrome associated with Acanthamoeba in-
fection and other opportunistic organisms. Arch Pathol
Lab Med 1986; 110:749-751.

24. Wiley CA, Safrin RE, Davis CE, Lampert PW, Braude AI,

Martinez AJ, Visvesvara GS. Acanthamoeba meningo-
encephalitis in a patient with AIDS. J Infect Dis 1987;
155:130-133.

25. Friedland LR, Raphael SA, Deutsch ES, Johal J, Martyn

LJ, Visvesvara GS, Lischner HW. Disseminated Acan-
thamoeba infection in a child with symptomatic human
immunodeficiency virus infection. Pediatr Infect Dis J
1992; ll:404--407.

26. Digregorio C. Acanthamoeba meningoencephalitis in . . .

AIDS. Arch Pathol Lab Med 1992; 116(12):1363-1365.

27. GuUett J, Mills J, Hadley K, Podemski B, Pitts L, Gelber

R. Disseminated granulomatous Acanthamoeba infec-
tion presenting as an unusual skin lesion. Am J Med
1979; 67:891-896.

28. Martinez AJ. Is Acanthamoeba encephalitis an opportu-

nistic infection? Neurology 1980; 30:567-574.

29. Lengy J, Jakovlzevich R, Tolls B. Recovery of hartmanel-

loid amoeba from a purulent ear discharge. Trop Dis
Bull 1971; 68:818.

Vol. 60 No. 4

PATHOGENIC FREE-LIVING AMEBAS— BOTTONE

269

30. Jadin JB. De la dispersion et du cycle des amibes libres.

Ann Soc Beige Med Trop 1974; 54:371-375.

31. Borochovitz D, Martinez AJ, Patterson GT. Osteomyelitis

of a bone graft of the mandible with Acanthamoeba cas-
tellanii infection. Human Pathol 1981; 12:573-576.

32. Centers for Disease Control. Acanthamoeba keratitis as-

sociated with contact lenses — United States. Morbidity
Mortality Weekly Rep 1986; 35:405-408.

33. Jones DB, Visvesvara GS, Robinson NR. Acanthamoeba

polyphaga keratitis and Acanthamoeba uveitis associ-
ated with fatal meningo-encephalitis. Trans Ophthal-
mol Soc U K 1975; 95:221-232.

34. Auran JD, Starr MB, Jakobiec FA. Acanthamoeba kera-

titis. A review of the literature. Cornea 1987; 6:2-26.

35. Asbell, PA. Acanthamoeba of the eye. Mt Sinai J Med

1993; 60:279-282.

36. Stehr-Green JK, Bailey TM, Brandt FH, Carr JH, Bond

WW, Visvesvara GS. Acanthamoeba keratitis in soft
contact lens wearers. J Am Assoc Med 1987; 258:57-60.

37. John T, Desai D, Sahm D. Adherence of Acanthamoeba

castellanii cysts and trophozoites to unworn soft contact
lenses. Am J Ophthalmol 1989; 108:658-664.

38. John T, Desai D, Sahm D. Adherence of Acanthamoeba

castellanii to new daily wear, extended wear, and dis-
posable soft contact lenses. CLAO J 1991; 17:109-113.

39. Bottone EJ, Madayag RM, Qureshi MN. Acanthamoeba

keratitis: synergy between amebic and bacterial cocon-
taminants in contact lens care systems as a prelude to
infection. J Clin Microbiol 1992; 30:2447-2450.

40. Donzis PB, Mondino BJ, Weissman BA, Bruckner DA.

Microbial contamination of contact lens care systems.
Am J Ophthalmol 1987; 104:325-333.

41. Larkin DFP, Livingston SK, Easty DL. Contamination of

contact lens storage cases by Acanthamoeba and bacte-
ria. Br J Ophthalmol 1990; 70:133-135.

42. Singh BN, Sayena V, Iyer SS. Production of viable sterile

cysts of free-living amoebae and role of bacteria on ex-
cystment. Indian J Exp Biol 1965; 3:110-112.

43. Moore MB, Ubelaker J, Silvany R, Martin J, McCulley JP.

Scanning electron microscopy of Acanthamoeba castel-
lanii: adherence to surfaces of new and used contact
lenses and to human corneal button epithelium. Rev
Infect Dis 1991; 13(Suppl 5):S423.

44. Morton LD, McLaughlin GL, Whitley HE. Adherence

characteristics of three strains of Acanthamoeba. Rev
Infect Dis 1991; 13(Suppl 5):S424.

45. Thong YH, Ferrante A. Migration patterns of pathogenic

and nonpathogenic Naegleria species. Infect Immun
1986; 51:177-180.

46. Martinez AJ, Markowitz SM, Duma RJ. Experimental

pneumonitis and encephalitis caused by Acanthamoeba
in mice: pathogenesis and ultrastructural features. J
Infect Dis 1975; 131:692-699.

47. Ferrante A, Bates EJ. Elastase in pathogenic free-living

amoebae Naegleria and Acanthamoeba species. Infect
Immun 1988; 56:3320-3321.

48. Yuguang H, Niederkorn JY, McCulley JP, Stewart GL,

Meyer DR, Silvany R, Dougherty J. In vivo and in vitro
coUagenolytic activity of Acanthamoeba castellanii. In-
vest Ophthal Visual Sci 1990; 31:2235-2240.

49. Michelson MK, Henderson WR Jr, Chi EY, Fritsche TR,

Klebanoff SJ. Ultrastructural studies on the effect of
tumor necrosis factor on the interaction of neutrophils
and Naegleria fowleri. Am J Trop Med Hyg 1990; 42:
225-233.

50. Senaldi G, Strosselli M, DiPerri G, Parisi A, Scaglia M,

Rondanelli EG. Naegleria fowleri: phase contrast cine-

micrographic study of phagocytosis of human erythro-
cytes. Exp Parasitol 1989; 69:290-293.

51. Marciano-Cabral FM, Patterson M, John DT, Bradley SG.

Cytopathogenicity of Naegleria fowleri and Naegleria
gruberi for established mammalian cell cultures. J Par-
asitol 1982; 68:1110-1116.

52. John DT, Cole TB Jr, Marciano-Cabral FM. Sucker-like

structures on the pathogenic amoeba Naegleria fowleri.
Appl Environ Microbiol 1984; 47:12-14.

53. Ferrante A, Rowan-Kelly B. Activation of the alternative

pathway of complement by Acanthamoeba culbertsoni.
Clin Exp Immunol 1983; 54:477-485.

54. Ferrante A. Immunity to Acanthamoeba. Rev Infect Dis

1991; 13(Suppl 5):S403-409.

55. Whiteman LY, Marciano-Cabral F. Resistance to highly

pathogenic Naegleria fowleri amoebae to complement-
mediated lysis. Infect Immun 1989; 57:3869-3875.

56. Reilly MF, Marciano-Cabral F, Bradley DW, Bradley SG.

Agglutination of Naegleria fowleri and Naegleria gru-
beri by antibodies in human serum. J Clin Microbiol
1983; 17:576-581.

57. Thong YH, Ferrante A, Sheperd C, Rowan-Kelly B. Re-

sistance of mice to Naegleria meningoencephalitis
transferable by immune serum. Trans R Soc Trop Med
Hyg 1978; 72:650-652.

58. Ferrante A, Thong YH. Antibody induced capping and

endocytosis of surface antigens in Naegleria fowleri. Int
J Parasitol 1979; 9:599-601.

59. Ferrante A, Thong YH. Unique phagocytic process in neu-

trophil killing of Naegleria fowleri. Immunol Lett 1980;
2:37^1.

60. Ferrante A, Carter RF, Lopez AF, Rowan-Kelly B, Hill

NL, Vadas MA. Depression of immunity to Naegleria
fowleri in mice by selective depletion of neutrophils
with monoclonal antibody. Infect Immun 1988; 56:
2286-2291.

61. Ferrante A, Mocatta TJ. Human neutrophils require ac-

tivation by mononuclear leucocyte conditioned medium
to kill the pathogenic free-living amoeba, Naegleria
fowleri. Clin Exp Immunol 1984; 56:559-566.

62. Ferrante A, Abell TJ. Conditioned medium from stimu-

lated mononuclear leukocytes augments human neu-
trophil-mediated killing of a virulent Acanthamoeba sp.
Infect Immun 1986; 51:607-617.

63. Ferrante A. Augmentation of the neutrophil response by

tumor necrosis factor alpha. Infect Immun 1989; 57:
3110-3115.

64. Ferrante A, Hill NL, Abell TJ, Pruul H. Role of myelo-

peroxidase in the killing of Naegleria fowleri by lym-
phokine-altered human neutrophils. Infect Immun
1987; 55:1047-1050.
64a. Qureshi MN, Bottone EJ. Acanthamoeba keratitis: a ra-
tional approach to microbiologic diagnosis. Med Micro-
biol Lett 1993;2:117-124.

65. Johns KJ, Head WS, Elliot JH, O'Day DM. Isolation and

identification of Acanthamoeba in corneal tissue. CLAO
J 1987; 13:272-276.

66. Epstein RJ, Wilson LA, Visvesvara GS, Plourde EG Jr.

Rapid diagnosis of Acanthamoeba keratitis from cor-
neal scrapings using indirect fluorescent antibody
staining. Arch Ophthalmol 1986; 104:1318-1321.

67. Wilhelmus KR, Osato MS, Robinson NM, Jones DB. Rapid

diagnosis of Acanthamoeba keratitis using calcofluor
white. Arch Ophthalmol 1986; 104:1309-1312.

68. Silvany RE, Luckenbach MW, Moore MB. The rapid de-

tection of Acanthamoeba in paraffin-embedded sections

270

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

of corneal tissue with calcofluor white. Arch Ophthal-
mol 1987; 1366-1367.

69. Marines HM, Osto MS, Font RL. The value of calcofluor

white in the diagnosis of mycotic and Acanthamoeba
infections of the eye and ocular adnexa. Ophthalmology
1987; 94:23-26.

70. Johns KJ, Head WS, Robinson RD, Williams TE, O'Day

DM. Examination of the contact lens with light micros-
copy: an aid in diagnosis of Acanthamoeba keratitis.
Rev Infect Dis 1991; 13(suppl 5):S425.

71. Visvesvara GS, Mirra SS, Brandt FH, Moss DM, Mathews

HM, Martinez AJ. Isolation of two strains of Acan-
thamoeba castellanii from human tissues and their
pathogenicity and isoenzyme profiles. J Clin Microbiol
1983; 18:1405-1412.

72. Epstein RJ, Wilson LA, Visvesvara GS, Plourde EG Jr.

Rapid diagnosis of Acanthamoeba keratitis from cor-
neal scrapings using indirect fluorescent antibody
staining. Arch Ophthalmol 1986; 104:1318-1321.

73. Robin JB, Chan R, Andersen BR. Rapid visualization of

Acanthamoeba using fluorescein-conjugated lectins.
Arch Ophthalmol 1988; 106:1273-1276.

74. Krogstad DJ, Visvesvara GS, Walls KW, Smith JW. Blood

and tissue protozoa. In: Balows A, Hausler WS, Jr,

Herrmann KL, Isenberg HD, Shadomy HJ, eds. Manual
of clinical microbiology, 5th ed. Washington, D.C.:
American Society for Microbiology 1991; 727-750.
74a. Qureshi MN, Perez AA II, Madayag RM, Bottone EJ.
Inhibition of Acanthamoeba species by Pseudomonas
aeruginosa: rationale for their selective exclusion in
corneal ulcers and contact lens care systems. J Clin
Microbiol 1993;31:1908-1910.

75. Gradus MS, Koenig SB, Hyndivk RA, DeCarlo J. Filter-

culture technique using amoeba saline transport me-
dium for the noninvasive diagnosis of Acanthamoeba
keratitis. Am J Clin Pathol 1989; 92:682-685.

76. Bottone EJ, Qureshi MN, Asbell PA. A simplified method

for demonstration and isolation of Acanthamoeba or-
ganisms from corneal scrapings and lens care systems.
Am J Ophthalmol 1992; 13:214-215.

77. Stopak SS, Roaf M, Nauheimer RC, Turgein PW, Sossi G,

Kowalski RP, Thaft RA. Growth of Acanthamoeba on
human corneal epithelial cells and keratocytes in vivo.
Invest Ophthalmol Visual Sci 1991; 32:354-359.

78. McLaughlin GL, Stimac JE, Luke SM, Kuhlenschmidt

MS, Vernon RA, Morton LD, Visvesvara GS, Whiteley
HE. Development of Acanthamoeba cornea coincuba-
tion assays. Rev Infect Dis 1991; 13(suppl 5):S397-398.

Free-Living Amebas:

Infection of the Central Nervous System

A. Julio Martinez, M.D.
Abstract

Pathogenic free-living amebas of the genera Naegleria and Acanthamoeba and the lepto-
myxid ameba of the order Leptomyxida may be capable of producing disease in the central
nervous system of human beings and animals. These amebas are distributed worldwide in
thermally polluted streams, coastal and fresh water, dust, soil, and sewage, and heating,
ventilating, and air conditioning units. N. fowleri may produce primary amebic menin-
goencephalitis, a rapidly fatal central nervous system infection. By contrast, the Acan-
thamoeba spp. and the recently described leptomyxid ameba may produce granulomatous
amebic encephalitis, a protracted central nervous system disease, usually in immunocom-
promised hosts. The leptomyxid ameba may produce clinical symptoms similar to and
histopathologic features almost identical to those in GAE. Contact-lens wearers may also
develop Acanthamoeba keratitis, chronic ulceration of the corneal epithelium due to Acan-
thamoeba spp. The various central nervous system diseases produced by these free-living
amebas result in divergent epidemiological patterns, diverse clinical manifestations, and
distinct pathological features, and require different treatment.

The story of free-living amebic infection has an
interesting beginning. In 1930 Sir Aldo Castel-
lani found and isolated a free-living, apparently
harmless ameba growing in a yeast culture (1). In
1957 amebas were found to produce cytopathic
effects on contaminated cell cultures (2). The dis-
ease caused by these amebas was first produced in
laboratory animals in 1958 by Clyde Culbertson
as a result of a fortuitous observation at the Eli
Lilly Laboratory (3). Polio vaccine was being pro-

Adapted from the author's presentation at the Workshops in
Clinical Lab Sciences Continuing Education Program, "Con-
temporary Protozoal Pathogens: Acanthamoeba, Naegleria,
Leptomyxid Ameba, and Microsporidia," at The Mount Sinai
School of Medicine, New York, NY, on October 25, 1991. Final
revision received March 1992. From the University of Pitts-
burgh School of Medicine and the Department of Pathology
(Neuropathology), Presbyterian University Hospital & Mon-
tefiore University Hospital, Pittsburgh, PA. Address reprint
requests to the author at the Division of Neuropathology,
Presbyterian University Hospital, DeSoto at O'Hara Street,
Pittsburgh, PA 15213.

Supported in part by the Pathology Education and Re-
search Foundation (PERF) of the Department of Pathology,
University of Pittsburgh.

duced, and each batch was tested for safety on
tissue culture plates. On one group of plates,
plaques suggesting live virus appeared, and mice
were inoculated with plate washings. The mice
died from a meningoencephalitis, and detection of
a possible nonattenuated virus was suspected. On
further study, unusual cells were found in the
brains of inoculated mice, and reexamination of
the plates revealed a free-living ameba. It was
identified as an Acanthamoeba spp., a common,
small, free-living ameba noted to occur as an air-
borne contaminant of cell cultures in other labo-
ratories (4). If not for the animal inoculations, the
plaques would have passed as simply another in-
stance of laboratory contamination.

The discovery that this presumably harmless
free-living ameba, even when simply instilled
into the nose, could invade the olfactory neuroep-
ithelium, migrate to the brain, multiply in the
subarachnoid space, invade the cerebral cortex,
and produce a rapidly fatal meningoencephalitis
was interesting and amazing. These important
findings suggested the possibility of infection in
humans and animals. These dramatic discoveries
led to recognition of cases in humans in 1965,

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

271

272

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

TABLE 1

Number of Cases Worldwide of Primary Amebic
Meningoencephalitis (PAM) and Granulomatous Amebic
Encephalitis (GAE) at September 1, 1991

Country

GAE

PAM

Total

Confirined*

Argentina

Australia

Belgium

Brazil

Canada

China

Colombia

Czechoslovakia

Great Britain

Honduras

India

Italy

Japan

Mexico

New Guinea

New Zealand

—

Nigeria

Panama

—

Peru

—

South Africa

—

South Korea

—

Thailand

—

United States

101

Venezuela

Zambia

—

Not confirmedt

Brazil

Cuba

Great Britain

Hungary

India

Mexico

Uganda

United States

Totals

156

229

* By autopsy, biopsy, or culture with isolation of amebas.
t Not confirmed by autopsy or culture of amebas.

when Fowler and Carter isolated N. fowleri from
cerebrospinal fluid and reported the first case,
from Adelaide, South Australia, and identified
the responsible ameba in 1965 (5); however, what
was originally described in this 1965 report as an
Acanthamoeba spp. turned out to be N. fowleri.
Cecyl Butt reported the first human case from
Florida in 1966 and called the disease primary
amebic meningoencephalitis (PAM) (6). In 1968
Callicott reported the first case from Virginia (7).
In 1969 Symmers reported two cases from mate-
rial kept in a museum in London; one of the two
patients, from Essex, northeast of London, died in
1909, and the other, from Belfast, Northern Ire-

land, died in 1937 (8). In 1970 J. dos Santos ret-
rospectively studied more than 400 cases of me-
ningoencephalitis in Richmond, and found five
additional cases, the earliest occurring in August
1937 (9).

There have been numerous reports of infec-
tion by N. fowleri (10-12) and by Acanthamoeba
spp. (13-23). Recently a new species of free-living
ameba, Balamuthia mandrillaris (23a), has been
isolated, chiefly in patients with acquired immu-
nodeficiency syndrome (AIDS) (24-28).

Classification and Protozoology

The free-living amebas comprise a group of
ubiquitous protozoa belonging to the genera Nae-
gleria (with a flagellate state) and Acanthamoeba
and an ameba of the order Leptomyxida. They
include several species of amphizoic amebas
pathogenic to humans and animals. Amebas of
the genus Naegleria are capable of producing a
rapidly fatal, generalized meningoencephalitis,
primary amebic meningoencephalitis (PAM).
About 156 cases of PAM have been reported
worldwide, 64 in the United States as of Septem-
ber 1, 1991 (Table 1). Acanthamoeba spp. and the
leptomyxid ameba may produce a multifocal en-
cephalitis, granulomatous amebic encephalitis
(GAE). About 73 cases of GAE have been reported
worldwide, 39 in the United States as of Septem-
ber 1, 1991 (Table 1). In addition, Acanthamoeba
spp. may produce acanthamoebic keratitis, a
chronic ulceration of the corneal epithelium. In
this presentation some historical and protozoo-
logic aspects, the clinical symptoms, diagnosis
and pathologic features of free-living amebic in-
fections will be discussed and compared (Table 2).

Naegleria. The amebas which have been iso-
lated from humans are morphologically identical
to the common, free-living A^. gruberi. The tropho-
zoites are active and usually elongated with
broadly rounded, granule-free processes called
lobopodia that erupt from the surface; granular
cytoplasm flows into them. The cytoplasm is
abundant with vacuoles. There is a conspicuous,
clear nuclear halo and a dense, spherical nucleo-
lus. When rounded, the trophozoite measures 10-
15 |jLm in diameter (Fig. 1). The cysts are usually
spherical, smooth, single-layered, and refractile.
They measure about 10 p-m in diameter. They
have one or two small flat pores. A'', australiensis
is pathogenic to mice. The other species of Nae-
gleria (N.jadini, N. gruberi, N. lovaniensis) have
not so far been found to be pathogenic to human
beings.

Vol. 60 No. 4

NEUROLOGIC INFECTION— MARTINEZ

273

Acanthamoeha. The trophozoites of Acan-
thamoeba spp. (A. castellanii, A. culbertsoni, A.
polyphaga, A. palestinensis) are recognized by
slender, spinelike processes called acanthopodia.
When rounded, the cells measure from 25 to 35
|jLm. Locomotion is by a slow, gliding movement.
The cytoplasm is finely granular and usually con-
tains a single nucleus which, like that of Naegle-
ria, has a large, dense, central nucleolus sur-
rounded by a clear nuclear halo. Water and food
vacuoles are usually evident in the cytoplasm.
The cysts are characterized by a double wall, are
star shaped, hexagonal, polygonal, or spherical,
and measure about 15 ixm in diameter. They have
two or more pores, opercula or ostioles.

Leptomyxid Ameba. The trophozoites and
cysts of the leptomyxid ameba are similar to those
of Acanthamoeha spp., having the same dimen-
sions and fine structural details (24).

A^. fowleri:
Primary Amebic
Meningoencephalitis

Pathogenesis and Epidemiology. The infec-
tion produced by A'^. fowleri is acquired by expo-
sure to contaminated water in which the thermo-
philic strain of the pathogenic ameba multiplies.

Most patients with PAM are healthy, normal
children or young adults; nearly all have fulmi-
nating illnesses rapidly leading to death; and the
vast majority had been swimming or were other-
wise exposed to stagnant water, or had inhaled
dust or soil containing amebic cysts (12).

The olfactory neuroepithelium is the portal of
entry and is the direct route of invasion by A'^.
fowleri. The incubation period may vary from two
to seven days in infected humans, and from 7 to
15 days in experimental animal infections. This,
of course, may depend on the size of the inocula
and the virulence of the amebas.

Swimming pools and artificial lakes are the
principal habitat of N. fowleri. Warm water dur-
ing the hot summer months, warm water near the
discharge outlet of power plants, and poorly chlo-
rinated warm swimming pools have been shown
to facilitate the growth of the pathogenic A^. fowl-
eri.

Clinical Signs and Symptoms. The onset of
clinical symptoms in PAM is usually abrupt and
rapidly progressive, and consists of headache, fe-
ver, pharyngitis, or nasal obstruction or dis-
charge. Occasionally an initial symptom is distor-
tion of smell and taste. Headache, vomiting, and
fever may persist, and then drowsiness, confu-

sion, and stiff neck develop. Convulsions may also
occur. Progressive deterioration follows, leading
to deep coma but with minimal, if any, focal neu-
rologic signs. The vast majority of patients die
one to two weeks after the first symptoms appear.

Differential Diagnosis and Techniques.
Free-living amebas should be suspected as etio-
logic agents in the differential diagnosis of every
case of purulent (suppurative) meningoencepha-
litis in which bacteria are not found. Examina-
tion of the fresh CSF is mandatory because it may
reveal motile amebas. The CSF of infected pa-
tients is cloudy and slightly hemorrhagic. It dem-
onstrates increased cellularity composed chiefly
of polymorphonuclear leukocytes, but without
bacterial pathogens, and protein concentration is
usually increased. Glucose content is normal or
decreased. Ameba motility can be detected as the
trophozoite undergoes alterations in shape and
configuration. With wet preparations, motile
amebas can be visualized in the CSF.

Smears of CSF may be stained with Wright
or preferably with Giemsa. These techniques may
reveal the characteristic trophozoites with sky-
blue cytoplasm and delicate pink nuclei. Lympho-
cytes and monocytes have larger nuclei and scant
cytoplasm. A history of good health in a child or
young adult and a recent history of water-related
sport activities are good clues for suspecting PAM
(12).

Pathological Features. The significant histo-
pathologic features in PAM have been confined to
the CNS. Macroscopically, the brain shows severe
edema characterized by narrowing of sulci and
flattening of gyri with some herniations. A
meningeal exudate is generally obvious in the
basilar cisterns, but is minimal or nonexistent
over the cerebral convexities, and is sometimes
associated with focal hemorrhages on the cerebral
cortex. The olfactory bulbs tend to be edematous,
soft, friable, and often adherent to the adjacent
cerebral cortex.

Microscopically, the purulent leptomeninge-
al exudate is composed of polymorphonuclear and
mononuclear leukocytes with a variable number
of amebic trophozoites (Fig. 2). The inflammatory
reaction extends to the cerebral cortex, which
shows variable degrees of inflammation, neuro-
nal necrosis, edema, hemorrhages, and invasion
with amebic trophozoites. The brainstem is also
affected by the meningoencephalitic process, but
shows less inflammatory change than do the ce-
rebral hemispheres. Necrotizing angiitis is occa-
sionally seen in PAM. Frequently, clusters of
amebic trophozoites without inflammatory reac-

274

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

TABLE 2

Comparison o/Naegleria and Acanthamoeha-Leptomyxid as Organisms Causative of Disease

Naegleria fowleri

Acanthamoeba-Leptomyxid

Protozoology

Epidemiology

Incubation period
Portal of entry
Onset

Spread into CNS
Organ affected
Clinical course
Signs and symptoms

CSF and lab data

Pathology, host
response, amebic
forms

Differential diagnosis

Therapy

Trophozoite: 10-15 (xm diameter
Round, conspicuous karyosome, clear

nuclear halo. Lobopodia.
Cyst: Round with pores.
Mitochondria: Dumbbell shape.
Divide by simple binary fission

Good health previously. Recent history of
swimming in lake or swimming pool.
Hot summer months. 156 cases
worldwide

2—7 days humans; 1-2 weeks animals

Olfactory neuroepithelium

Acute, fast

Direct; amyelinic nervous plexus
Brain

Acute, fulminant

Severe headache; anorexia, nausea,
vomiting; fever (39^1°C), ataxia,
diplopia, stiff neck, coma

CSF similar to bacterial meningitis but
sterile. Neutrophilic pleocytosis. High
protein. Low glucose. Direct
examination of fresh CSF: trophozoites
active and mobile. Cultivation-
inoculation in mice

Acute purulent leptomeningitis.
Hemorrhagic necrotizing
meningoencephalitis. Marked brain
edema. Perivascular collection of
amebas. Only trophozoites in brain

No distinct diagnostic differences from
acute pyogenic (bacterial) meningitis

Amphotericin B

Trophozoite: 25-35 |xm diam.
Mitochondria: Oval. Round, central

karyosome, with clear nuclear halo.

Acanthopodia.
Cyst: Star shape with double walls.

History of poor health.

Immunological incompetent patient AIDS.

No history of swimming. 73 cases

worldwide.

Uncertain >10 days

Lung, skin, eyes

Chronic, slow and insidious

Hematogenous

Lung, brain, eyes

Chronic, usually prolonged

Ataxia, cranial nerve palsies; mental
abnormalities; hemiparesis; meningism;
coma

Consistent with viral

Chronic granulomatous encephalitis with
focal necrosis. Cysts and trophozoites in
brain

No distinct diagnostic differences from
acute/chronic pyogenic (bacterial)
meningitis, TB, fungal or viral
encephalitis. Ocular: Herpes simplex or
fungal keratitis

Ketoconazol (Miconazol), sulfadiazine,
propamidine isethionate (Brolene)

tion can be seen in the CNS. The amebic tropho-
zoites follow a centripetal course of invasion, from
cerebral cortex to the deep subcortical white mat-
ter. No amebic cysts are present in the CNS le-
sions.

Treatment and Prognosis. Except for sup-
portive measures, such as control of temperature,
fluid, and electrolytes, specific therapy in most
cases has little influence on the natural course of
PAM. Amphotericin B is the drug of choice for N.
fowleri infections, but is of value only when given
early in the course of the illness. Both amphoteri-
cin B and miconazole have been shown to be ef-
fective in vitro against Naegleria species.

Acanthamoeba spp. and

Leptomyxid Amebas
Granulomatous Amebic
Encephalitis

The majority of patients with GAE are chron-
ically ill, immunocompromised, and debilitated
with other diseases or are undergoing immuno-
suppressive therapy and have no history of recent
water-sports activities. Additional cases associ-
ated with AIDS should be expected (11-28).

Pathogenesis and Epidemiology. In GAE the
route of invasion and penetration into the CNS
appears to be hematogenous, probably from a pri-
mary focus in the respiratory tract (lung) or skin.

Vol. 60 No. 4

NEUROLOGIC INFECTION— MARTINEZ

275

However, the amebas may also penetrate directly
into the olfactory neuroepithelium. The lower re-
spiratory tract, skin (skin ulcers), and eyes (cor-
neal ulcers) appear to be the sites of primary le-
sions in patients who have Acanthamoeba spp.
and leptomyxid ameba infections. Experimen-
tally, an amebic pneumonitis due to Acan-
thamoeba spp. has been reported.

The infection produced by Acanthamoeba
spp. and the leptomyxid ameba are probably op-
portunistic in the majority of cases. Patients with
AIDS are particularly vulnerable. Acanthamoeba
spp. and the leptomyxid ameba are also wide-
spread protozoa that can be found as normal flora
in healthy individuals and can be isolated from
samples of fresh water, air, soil, bottled mineral
water, and even from dialysis machines and air
conditioning units.

The incubation period is probably more than
10 days and the clinical course is subacute and
chronic, usually protracted, lasting several weeks.

Clinical Signs and Symptoms. The clinical
symptoms are mental abnormalities and menin-
gism with localizing neurological signs, and even-
tually coma and death.

Differential Diagnosis and Techniques. Ret-
rospective diagnosis of free-living amebic infec-
tions has been made by studying microscopic
slides from CNS tissues. In such instances iden-
tification of the responsible amebas has rested
chiefly on immunologic studies of either sera or
postmortem tissues, on morphologic features such
as size and shape of trophozoites, and on the
shapes of the cysts as noted in formalin-fixed and
paraffin-embedded tissue. Immunofluorescent-
antibody and immunoperoxidase techniques have
been used, as well as electron microscopy.

The distinction between PAM and GAE is not
difficult, especially in histologic preparations of
infected tissues. In tissues the trophozoites of
Naegleria measure approximately 10-15 ixm,
whereas those of Acanthamoeba are larger. Cysts
with a wrinkled double wall in tissue are found
only in infections with Acanthamoeba spp. and
the leptomyxid ameba.

A distinctive feature of N. fowleri is its abil-
ity to convert into a flagellate form which can be
induced by diluting the culture with water. In
CSF, the amebas are easily confused with leuko-
cytes but can be identified using an unstained wet
preparation. By this method, the trophozoites can
be seen to contain a single central or eccentric
nucleus with a conspicuous karyosome and char-
acteristic lobopodia.

Brain biopsy may be done for a diagnosis.

particularly when a brain abscess, brain tumor or
any other space-occupying mass, or a viral en-
cephalitis is suspected.

Pathologic Features: Central Nervous Sys-
tem. The pathologic lesions in GAE may be
present in the brain, the eyes, the lungs, or the
skin. The cerebral hemispheres are usually char-
acterized by mild, focal edema with areas of soft-
ening associated with necrosis and recent hemor-
rhage. There is flattening of the gyri and
narrowing of the sulci only in the affected areas.
The rest of the cerebral cortex is unremarkable.
Bilateral uncal notching and cerebellar tonsillar
herniation may be present, but they are incon-
spicuous. The posterior fossa structures, dien-
cephalon, thalamus, and brainstem are usually
the most affected areas. On CT scan the lesions
may resemble a space-occupying mass, brain tu-
mor, brain abscess, or hemorrhagic infarct.

The leptomeninges, mainly over the most af-
fected cortical areas, are opaque and might reveal
a moderate amount of purulent exudate and vas-
cular congestion. The rest of the leptomeninges
are transparent with normal vascularity. The ol-
factory bulbs and spinal cord are spared. The
modest chronic inflammatory exudate covering
the cortical gray matter is composed of lympho-
cytes, monocytes, plasma cells, histiocytic ele-
ments, and few, if any, polymorphonuclear leuko-
cytes.

Angiitis with necrotizing arteritis and fibrin-
oid necrosis may be seen. The invading amebic
trophozoites and cysts take a centrifugal course,
usually from deep gray or white matter areas to
the brain surface.

Trophozoites and cysts are the amebic forms
observed, scattered throughout the lesions, but
preferentially located in perivascular spaces and
invading blood vessel walls (Fig. 3).

The majority of the lesions of GAE are char-
acterized by a chronic and granulomatous reac-
tion, with multinucleated giant cells. Some pa-
tients under immunosuppression may not have
granulomatous inflammation because of the im-
paired immune system.

Pathologic Features: Eyes. In patients who
have Acanthamoeba keratitis, the acanthamoebic
trophozoites and cysts may also be identified in
direct smears from the surface of the corneal ul-
cerations, and may be cultured from these mate-
rials.

Pathologic Features: Lungs. Amebic tropho-
zoites and cysts may be present within the pul-
monary alveoli encompassed by modest mononu-
clear cell infiltrate. These features are

276

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

Fig. 1. Trophozoite of Naegleria fowleri within brain of ex-
perimentally infected mouse, inoculated intranasally with
10,000 trophozoites per cubic mL. Trophozoite is in close con-
tact with blood vessel. Dense nucleolus is encompassed by
finely granular nuclear chromatin. Cytoplasm contains empty
vacuoles, lysosomes, mitochondria, and engulfed dense axo-
plasmic nests (electron micrograph; original magnification
x5,000).

characteristic of acanthamoebic pneumonitis.
From the lower respiratory tract, amebic tropho-
zoites and cysts may reach the CNS by hematog-
enous spread.

Pathologic Features: Skin. Among the re-
ported cases of GAE, several patients had ulcer-
ations or nodular lesions of the skin in which
Acanthamoeba spp. were found on skin biopsies.
This finding suggests a possible portal of entry of
the protozoa which can reach the CNS by the he-
matogenous route. The skin lesions, particularly
if not ulcerated, may also represent a terminal
hematogenous dissemination of the protozoa,
forming multifocal subcutaneous inflammatory
foci and abscesses without disruption of the epi-
dermis.

Treatment and Prognosis. Ketoconazole and
clotrimazole are effective against some strains of
Acanthamoeba spp. in vitro. Propamidine isethi-
onate has been used with success to treat acan-
thamoebic keratitis. Sulfadiazine appears to be
less active against Acanthamoeba spp.

Summary

The prognosis for patients who have the neu-
rological disease produced by free-living amebas
is generally poor. Effective management of PAM
and GAE will depend on the development of cu-
rative drugs, on increased awareness of free-liv-
ing amebic infections, and on methods of early
diagnosis.

Fig. 2. Trophozoites of Naegleria fowleri within a Virchow-Robin space in patient with PAM. No accompa-
nying inflammatory exudate. Neuropil is edematous and neurons are shrunken and ischemic (H&E; original
magnification 400 x ).

Vol. 60 No. 4

NEUROLOGIC INFECTION— MARTINEZ

277

Fig. 3. Trophozoites and cysts of Acanthamoeba castellanii within area of necrotizing hemorrhagic enceph-
alitis in occipital lobe in patient with GAE (H&E; original magnification 400 x ).

References

1. Castellani A. An amoeba found in culture of a yeast: pre-

liminary note. J Trop Med Hyg June 2, 1930; p 160.
Second Note: July 1, 1930. Third Note: August 1, 1930.

2. Johnes WG, Fullmer HM, Li CP. Free-living amoebae as

contaminants in monkey kidney tissue cultures. Proc
Soc Exp Biol Med 1957; 96:484.

3. Culbertson CG, Smith JW, Minner JR. Acanthamoeba:

observations on animal pathogenicity. Science 1958;
127:1506.

4. Culbertson CG. Pathogenic Acanthamoeba (Hartman-

nella). Am J Clin Pathol 1961; 35:195-202.

5. Fowler M, Carter RF. Acute pyogenic meningitis probably

due to Acanthamoeba spp: a preliminary report. Br Med
J 1965; 2:240-242.

6. Butt CG. Primary amebic meningoencephalitis. N Engl J

Med 1966; 274:1473-1476.

7. Callicott JH Jr. Amebic meningoencephalitis due to free-

living amebas of the Hartmannella iAcanthamoeba)-
Naegleria group. Am J Clin Pathol 1968; 49:84-91.

8. Symmers W St C. Primary amoebic meningoencephalitis

in Britain. Br Med J 1969; 4:449^54.

9. Dos Santos JG. Fatal primary amebic meningoencephali-

tis: a retrospective study in Richmond, Virginia. Am J
Clin Pathol 1970; 54:737.

10. Visvesvara GS, Stehr-Green JK: Epidemiology of free-liv-

ing ameba infections. J Protozool 1990; 37:255-335.

11. Ma P, Visvesvara GS, Martinez AJ, Frederick HT, Pierre-

Marc D, Samyer TK. Naegleria and Acanthamoeba in-
fections: review. Rev Infect Dis 1990; 12:490-513.

12. Martinez AJ, Santos Neto JG, Nelson EC, Stamm EP,

Willaert E. Primary amoebic meningoencephalitis.
Pathol Annu 1977; 12:225-250.

13. Anzil AP, Rao Ch, Wrzolek MA, Visvesvara GS, Sher JH,

Kozlowski PB. Amebic meningoencephalitis in a pa-

tient with AIDS caused by a newly recognized opportu-
nistic pathogen leptomyxid ameba. Arch Pathol Lab
Med 1991; 115:21.

14. Carter RF, CuUity GJ, Ojeda VJ, Silberstein P, Willaert

E. A fatal case of meningoencephalitis due to a free-
living amoeba of uncertain identity — probably Acan-
thamoeba species. Pathology 1981; 13:51.

15. Cleland PG, Lawande RV, Onyamelukwe G, Whittle HC.

Chronic amebic meningoencephalitis. Arch Neurol
1982; 39:56.

16. Duma RJ, Helwig WB, Martinez AJ. Meningoencephalitis

and brain abscess due to a free-living amoeba. Ann In-
tern Med 1978; 88:468-473.

17. Gonzalez MM, Gould E, Dickinson G, Martinez AJ, Vis-

vesvara G, Cleary TJ, Hensley GT. Acquired immuno-
deficiency syndrome associated with Acanthamoeba in-
fection and other opportunistic organisms. Arch Pathol
Lab Med 1986; 110:749-751.

18. Gonzalez-Alfonso JE, Martinez AJ, Garcia V, Garcia-

Tamayo J, Cespedes G. Granulomatous encephalitis
due to a Leptomyxid amoeba. Trans Roy Soc Trop Med
Hyg 1991; 85:480.

19. Grunnet ML, Cannon GH, Kushner JP. Fulminant ame-

bic meningoencephalitis due to Acanthamoeba. Neurol-
ogy 1981; 31:174-178.

20. GuUett J, Mills J, Hadley K, Podemski B, Pitts L, Gelber

R. Disseminated granulomatous Acanthamoeba infec-
tion presenting as an unusual skin lesion. Am J Med
1979; 67:891-895.

21. Martinez AJ, Garcia CA, Halks-Miller M, Arce-Vela R.

Granulomatous amebic encephalitis presenting as a ce-
rebral mass lesion. Acta Neuropathol (Berl) 1980; 51:
85-91.

22. Martinez AJ. Is acanthamoebic encephalitis an opportu-

nistic infection? Neurology 1980; 30:567-574.

278

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

23. Martinez AJ. Acanthamoebiasis and immunosuppression.

Case report. J Neuropathol Exp Neurol 1982; 41:548-
557.

23a. Visvesvara GS, Schuster PL, Martinez AJ. Balmuthia
mandrillaris. N. G., N. Sp., agent of amebic meningo-
encephalitis in humans and other animals. J Euk Mi-
crobiol 1993;40(4 1:504-514.

24. Visvesvara GS, Martinez AJ, Schuster FL, Leitch GJ,

Wallace SV, Sawyer TK, Anderson M. Leptomyxid
ameba, a new agent of amebic meningoencephalitis in
humans and animals. J Clin Microbiol 1990; 28:2750-
2756.

24a. Wiley CA, Safrin RE, Davis CE, Lampert PW, Braude
AI, Martinez AJ, Visvesvara GS. Acanthamoeba me-
ningoencephalitis in a patient with AIDS. J Infect Dis
1987; 155:130-133.

25. Jaramillo-Rodriguez Y, Ch^vez-Macias LG, Olvera-Ra-

biela JE, Martinez AJ. Encephalitis por una nueva
amiba de vida libre, probablemente Leptomyxid. Pato-
logia (Mexico) 1989; 27:137-141.

26. Gardner HAR, Martinez AJ, Visvesvara GS, Sotrel A.

Granulomatous amebic encephalitis in an AIDS pa-
tient. Neurology 1991; 41:1993-1995.

27. DiGregorio C et al. Acanthamoeba meningoencephalitis

in a patient with acquired immunodeficiency syndrome:
first report from Europe. Arch Pathol Lab Med 1992;
116a2):1363-1365.

28. Friedland LR, Raphael SA, Deutsch ES, et al. Dissemi-

nated Acanthamoeba infection in a child with symptom-
atic human immunodeficiency virus infection. Pediatr
Infect Dis J 1992; ll(5):404-407.

Acanthamoeba Keratitis: There and

Back Again

Penny A. Asbell, M.D.

Acanthamoeba keratitis is cause for serious con-
cern to ophthalmologists. Its diagnosis and treat-
ment are difficult. Moreover, the infection can
cause permanent impairment of vision, and even
blindness.

Corneal trauma is a precondition for ocular
acanthamoebal infection. Even minor trauma
caused during the insertion, removal, or normal
wearing of contact lenses can provide an avenue
for entry of the microorganisms. Whatever the
environmental sources of acanthamoebae, which
are numerous, the organisms enter the cornea
and, depending on their number and virulence,
produce clinical infection.

In 1973 Visvesvara reported the first case of
Acanthamoeba infection of the eye (1). Few re-
ports of ocular acanthamoebal infection followed;
in 1982, not a single case of Acanthamoeba kera-
titis was found in a 30-year review of 700 patients
with corneal ulcers, all of whom had undergone
laboratory evaluation (2). Moreover, this study
did not point to contact lenses as a risk factor. It
was not until about 1985, after contact lens use
had dramatically increased, that the direct asso-
ciation between contact lens wear and Acan-
thamoeba keratitis was recognized (3, 4).

Diagnosis

The common clinical signs and symptoms of
ocular acanthamoebal infection are breakdown of

Adapted from the author's presentation at the Workshops in
CHnical Lab Sciences Continuing Education Program, "Con-
temporary Protozoal Pathogens: Acanthamoeba, Naegleria,
Leptomyxid Ameba, and Microsporidia," at The Mount Sinai
School of Medicine, New York, NY, on October 25, 1991. Final
revision received March 1993. From the Department of Oph-
thalmology, The Mount Sinai Medical Center. Address reprint
requests to the author at Box 1183, Mount Sinai Medical Cen-
ter, One Gustave L. Levy Place, New York, NY 10029.

the corneal epithelium, the presence of satellite
lesions, and iritis, sometimes with hypopyon. Of-
ten, after an initial period of virulence, the infec-
tion progresses slowly. Diagnosis is complicated
by the fact that the infection waxes and wanes,
with corneal healing followed by recurrent epi-
thelial deterioration. There may also be elevated
intraocular pressure and scleral inflammation.

The classic initial sign of acanthamoebal in-
fection that has penetrated deep into the cornea is
the combination of a central or paracentral ring
infiltrate and remarkably intense pain. Although
the degree of pain seems disproportionate to the
clinical signs, it becomes less surprising when one
considers that the cornea has the highest density
of free nerve endings in the body.

One method of diagnosis is to take corneal
scrapings for culture and microscopic examina-
tion. Special staining and culture techniques can
be useful. In some cases, biopsy may be necessary
to obtain a firm diagnosis. In any event, it is most
important, as Auran (5) pointed out, that speci-
mens be examined by someone who is thoroughly
familiar with the appearance of acanthamoebae.

Treatment

Initially, most patients required surgical
treatment, the most drastic procedure being cor-
neal transplantation. For example, Cohen et al.
reported, in 1985, two cases of Acanthamoeba
keratitis in which corneal transplantation was
necessary for both diagnosis and treatment (6),
and, in 1987, five cases treated by penetrating
keratoplasty (7).

Wright reported the first successful medical
treatment of Acanthamoeba keratitis (8). He used
neomycin and Broline, a combination of dibromo-
propamidine and propamidine isethionate, which

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

279

280

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

is available in England but not in the United
States. Treatment was prolonged but curative.

By now, many other drugs have been used in
efforts to cure ocular acanthamoebal infection.
Corticosteroids, in particular, have generated
controversy. In their 1991 study of the goals and
risks of corticosteroid use in treating different
forms of keratitis, Stern and Buttross concluded
that these drugs are "relatively contraindicated"
for use against Acanthamoeba keratitis (9).

Current data suggest that cure is possible if
Acanthamoeba keratitis is diagnosed early, if ap-
propriate medical treatment is begun promptly,
and if that treatment is continued for six to
twelve months. Two recently recommended drugs
are miconazole and ketoconazole, both antifungal
agents that can be used topically and systemi-
cally. In 1992, Larkin, Kilvington, and Dart suc-
cessfully used polyhexamethylene biguanide to
treat five of six patients with Acanthamoeba
keratitis that had proven resistant to conven-
tional antiamebic agents (10).

Risk Factors for
Acanthamoeba Keratitis

Lens Materials. Infection was not associated
with use of the first contact lenses, which were
made of polymethylmethacrylate. However, their
impermeability to oxygen did cause other prob-
lems. All modern contact lenses contain a certain
amount of water, which serves as the medium for
oxygen flux (11). The water content of soft hydro-
gel lenses is 50%-75%; that of rigid gas-perme-
able lenses is usually about 10%. In proportion to
their water content, lenses absorb material, in-
cluding potential pathogens, from cleaning solu-
tions, lens cases, and patients' hands (12, 13).
Thus, lens material itself can be a risk factor for
acanthamoebal infection.

Lens Cleaning and Disinfection. A soft con-
tact lens is routinely cleaned by placing it in the
palm of one's hand and gently rubbing it with a
few drops of a surfactant. In addition, soft lenses
are usually treated about once a week with en-
zymes to remove protein deposits. After both of
these procedures, the lenses should be disinfected
either by heating them in a unit designed for that
purpose or soaking them in hydrogen peroxide or
other chemical disinfectants. For patients with
daily-wear lenses, disinfection is usually an over-
night process after which the lenses are ready for
use.

Ten years ago, the preservatives in contact-
lens solutions were seen as major culprits in caus-
ing ocular redness, itching, and lens discomfort.

as well as corneal infiltrates. The prevalence of
these problems led practitioners to promote
the use of unpreserved solutions. Patients were
instructed to mix salt tablets and either tap water
or distilled water in place of preserved solutions
for lens cleaning and storage.

We did not know then that acanthamoebae
are ubiquitous in the environment; in particular,
we did not know that acanthamoebae are com-
monly present in the tapwater patients used to
make saline solution. The marked increase in
acanthamoebal infections during the 1980s cer-
tainly was related, in some degree, to patients'
switch from using preserved solutions that could
disinfect lenses to using unpreserved saline that
could not. The use of homemade saline is now
actively discouraged.

A related source of lens contamination and,
therefore, another risk factor for acanthamoebal
infection is many patients' reluctance to take
their lenses from a disinfecting solution and di-
rectly insert them in the eyes. After all, these
solutions are made of chemicals — they are not
"natural." As a result, some patients still rinse
their lenses with tapwater, again exposing them
to acanthamoebae as well as to hosts of other po-
tentially contaminating microorganisms. Other
patients, mistaking saline for a disinfectant, use
it alone instead of the regimen. Another potential
source of pathogens is contaminated eye drops,
which patients use to wet their eyes to increase
the comfort of lens wear.

Lens Storage Cases. Contact lens care sys-
tems are not intended to sterilize lenses; rather,
they are disinfection systems designed to reduce
the number of organisms that can adhere to con-
tact lenses. However, if these systems are not
used in accordance with manufacturers' recom-
mendations, lens disinfection will be inadequate.
This will be true, for example, if patients do not
use a surfactant before disinfecting their lenses,
do not keep the lenses in disinfectant for the pre-
scribed time, or rinse the lenses with tapwater
before insertion.

Until recently, ophthalmologists did not con-
sider the possibility that lens cases could be res-
ervoirs for potentially infective organisms. Then
Larkin, Kilvington, and Easty (12) examined 102
asymptomatic contact-lens wearers and their lens
cases. These patients had good vision, healthy
eyes, and unblemished lenses. But cultures of
their lens cases showed that nearly half contained
significant amounts of bacteria. Seven lens cases
contained acanthamoebae.

Donzis et al. (13) studied ten patients whose
lens cases were positive for acanthamoebae. All

Vol. 60 No. 4

ACANTHAMOEBA KERATITIS— ASBELL

281

ten cases were positive for gram-negative organ-
isms; in seven, these organisms included Pseu-
domonas species. Five cases also contained gram-
positive organisms, particularly bacilli, and six
contained fungi. These findings support the con-
clusion that acanthamoebae do not often occur
alone. It is possible that a synergistic or symbiotic
relationship exists in which other microorgan-
isms, particularly gram-negative ones, support
the growth of acanthamoebae to a density at
which they are pathogenic to the eye (13a-14).

Both of these studies made it obvious that
many patients' lens-care systems are signifi-
cantly contaminated. The problem, then, is not
just casual exposure of patients' lenses to acan-
thamoebae. Instead, patients' lenses are fre-
quently, even regularly, exposed to large, diverse,
and only partially defined populations of poten-
tially harmful organisms, including, of course,
acanthamoebae.

How, then, can systems for the care of contact
lenses be made free of acanthamoeba? Even if
that question could be satisfactorily answered,
there would remain another confounding factor:
when patients are instructed to leave their lenses
in a disinfecting solution for four hours, some will
cut the time to a scant thirty minutes.

Extended Wear. In 1989, Schein et al. (15)
found that soft-lens wearers who slept with lenses
on their eyes were at approximately ten to fifteen
times greater risk of developing ulcerative kera-
titis than were patients who wore their lenses
only during waking hours. In conducting this
study, the authors had to control for a wide range
of lens-wear patterns. They found, for example,
that patients who are instructed to wear their
lenses only during the day sometimes wear them
to sleep, whereas patients with extended-wear
lenses sometimes remove their lenses before go-
ing to sleep. Nonetheless, the study results dem-
onstrated that extended wear of contact lenses is
the single greatest risk factor for corneal ulcers in
soft-lens users. In a survey of New England oph-
thalmologists, Poggio et al. (16) estimated that
the incidence of ulcerative keratitis was only 4.1
per 10,000 daily wearers of soft lenses but 20.9
per 10,000 users of extended-wear soft lenses.
Nonetheless, these findings do not wholly explain
why the incidence of corneal ulcers, including
acanthamoebal ulcers, has increased so markedly
in recent years.

Lens Degradation. Repeated and prolonged
heating of soft lenses causes them to degrade. In
addition, whether they are used for extended
wear or are removed, cleaned, and disinfected
daily, contact lenses eventually develop deposits.

This is unavoidable, for the deposits are, in fact,
part of a biofilm. It can be argued that the biofilm
is needed, that producing it is the body's reaction
to the presence of foreign bodies — in this case,
contact lenses. According to this contention, the
lenses must be coated with secretions so that the
body can recognize the lenses and not react ad-
versely to them. However, the biofilm that coats
contact lenses also contains bacteria and contam-
inating detritus of various kinds, which act as
irritants and add to the risk of infection.

Disposable Lenses. Set against the back-
ground of all these closely related risk factors, the
development of disposable contact lenses seemed
to hold great promise. It seemed that these lenses
could eliminate all the difficulties relating to lens
disinfection, contamination of lens cases, and lens
degradation. The patient would wear a pair of
new, sterile lenses continually for one week,
throw them away, and then start the next week
with another pair of lenses. Perhaps there would
be no more Acanthamoeba keratitis.

This hope seemed to be borne out by early
experience. However, it was not long before an
association between the wearing of disposable
lenses and corneal infections was noted. Among
other infective organisms, acanthamoebae were
also back (17).

Conclusion

One obvious reason for the increase in Acan-
thamoeba keratitis is the even greater increase in
the number of people who wear contact lenses,
especially soft lenses. In one study (18) it was
found that 80% of the reported cases of this infec-
tion occurred among contact-lens wearers. Sev-
enty-five percent of these patients used daily-
wear or extended-wear soft hydrogel lenses.

Another reason for the resurgence of Acan-
thamoeba keratitis among contact-lens wearers is
indicated by the results of our recent study of 100
patients consecutively fitted with disposable
lenses (19). Ninety percent of these patients com-
plied with instructions to wear a pair of lenses for
no longer than one to two weeks. To our surprise,
however, 60% of the patients were removing and
reinserting their lenses at some point within the
prescribed period of wear, usually because of dust,
dirt, or ocular irritation. Virtually all of these pa-
tients reinserted their lenses without disinfecting
them; they simply rinsed the lenses, usually with
saline or tapwater. It was an old problem — the
one that made disposable lenses seem such a boon
in the first place.

The final answer is that, in all likelihood.

282

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

contact-lens wearers will always be at risk of oc-
ular acanthamoebal infection. That conclusion is
inherent in the ubiquity of these organisms, the
nature of contact-lens wear and care, and the
variability of human behavior. All of those factors
are emphatic reminders of the importance of ed-
ucating patients about the risk of infection and
the consequent need for rigorous adherence to
lens-care regimens. These measures, together
with regular, thorough follow-up of all contact-
lens patients, remain the first and most essential
defense against Acanthamoeba keratitis.

References

1. Jones CB, Robinson NR, Visvesvara GS. Paper presented

at the Ocular Microbiology and Immunology Group
Meeting, Dallas, Texas, September 1973. Cited in Jones
DB, Visvesvara GS, Robinson NR. Acanthamoeba poly-
phaga keratitis and Acanthamoeba uveitis associated
with fatal meningoencephalitis. Trans Ophthalmol Soc
UK 1987; 95:221-232.

2. Asbell PA, Stenson S. Ulcerative keratitis: survey of 30

years' laboratory experience. Arch Ophthalmol 1982;
100:77-80.

3. Moore MB, McCuUey JP, Luckenbach MD, Gelender H,

Newton C, McDonald MB, Visvesvara GS. Acan-
thamoeba keratitis associated with soft contact lenses.
Am J Ophthalmol 1985; 100:396-403.

4. CDC: Acanthamoeba keratitis associates with contact

lenses— United States. MMWR 1986; 35:405-408.

5. Auran JD, Starr MB, Jakobiec FA. Acanthamoeba kera-

titis: a review of the literature. Cornea 1987; 6:2-26.

6. Cohen EJ, Buchanan HW, Laughrea PA, et al. Diagnosis

and management of Acanthamoeba keratitis. Am J
Ophthalmol 1985; 100(3):389-395.

7. Cohen EJ, Parlato CJ, Arentsen JJ, Genvert GI, Eagle RC

Jr, Wieland MR, Laibson PR. Medical and surgical
treatment of Acanthamoeba keratitis. Am J Ophthal-
mol 1987; 103(5):615-625.

8. Wright P, Warhurst D, Jones BR. Acanthamoeba keratitis

successfully treated medically. Br J Ophthalmol 1985;
69(10):778-782.

9. Stern GA, Buttross M. Use of corticosteroids in combina-
tion with antimicrobial drugs in the treatment of infec-
tious corneal disease. Ophthalmology 1991; 98(6):847-
853.

10. Larkin DFP, Kilvington S, Dart JK: Treatment of Acan-

thamoeba keratitis with polyhexamethylene biguanide.
Ophthalmology 1992; 99(2):185-191.

11. Lippman JL. Contact lens materials: a critical review.

CLAO J 1990; 16:287-291.

12. Larkin DFP, Kilvington S, Easty DL. Contamination of

contact lens cases by Acanthamoeba and bacteria. Br J
Ophthalmol 1990; 74:133-135.

13. Donzis PB, Mondino BJ, Weissman BA, Bruckner DA.

Microbial analysis of contact lens care systems contam-
inated with Acanthamoeba. Am J Ophthalmol 1989;
108:53-56.

13a. Bottone EJ, Qureshi MN, Asbell PA. A simplified
method for demonstration and isolation of Acan-
thamoeba organisms from corneal scrapings and lens-
care systems. Am J Ophthalmol 1992; 113:214-215.

14. Bottone EJ, Madayag RM, Qureshi MN. Acanthamoeba

keratitis: synergy between amebic and bacterial cocon-
taminants in contact lens care systems as a prelude to
infection. J Clin Microbiol 1992; 30:2447-2450.

15. Schein OD, Glynn RJ, Poggio EC, et al. The relative risk

of ulcerative keratitis among users of daily-wear and
extended-wear soft contact lenses. A case-control study.
New Engl J Med 1989; 321(12):773-778.

16. Poggio EC, Glynn RJ, Schein OD, Seddon JM, Shannon

MJ, Shannon MJ, Scardino VA, Kenyon KR. The inci-
dence of ulcerative keratitis among users of daily-wear
and extended-wear soft contact lenses. N Engl J Med
1989; 321(12):779-783.

17. Picker L, Hunter P, Seal D, Wright P. Acanthamoeba

keratitis occurring with disposable contact lens wear.
Am J Ophthalmol 1989; 108:453.

18. Moore MB, McCuUey JP, Newton C, Cobo LM, Foulks GN,

et al. Acanthamoeba keratitis a growing problem in soft
and hard contact lens wearers. Ophthalmology 1987;
94:1645.

19. Asbell PA, Dunn MJ, Torres MA, Wang G, Starer KL.

Compliance in the care of disposable contact lenses: the
effect of patients' health beliefs. CLAO J 1993; 19:150-
152.

Epidemiology of Infections with
Free-Living Amebas and Laboratory
Diagnosis of Microsporidiosis

GOVINDA S. ViSVESVARA, Ph.D.

Infections Caused by
Free-Living Amebas
Free-living amebas occur worldwide. They have
been isolated from diverse environmental
sources, including fresh and frozen water; vege-
tables; heating, ventilating, and air conditioning
units; cooling towers of electric and nuclear power
plants; sewage; medicinal pools; dental treatment
units; dialysis units; bacterial, fungal, and mam-
malian cell cultures; contact lens solutions; cor-
neal scrapings; skin lesions; and the central ner-
vous system of humans (1-3). Among the
hundreds of free-living amebas that exist in na-
ture, only those belonging to two genera, Naegle-
ria and Acanthamoeba, and the newly discovered
leptomyxid ameba are known to cause disease in
humans (1-4).

Only one species of Naegleria, N. fowleri,
causes primary amebic meningoencephalitis
(PAM), which is an acute, hemorrhagic, necrotiz-
ing meningoencephalitis that usually results in
death within 5 to 10 days. Amebic trophozoites,
but not cysts, are seen in the brain tissue of pa-
tients who have died of PAM (Fig. 1, 2). Several
species o{ Acanthamoeba, on the other hand, are
known to cause granulomatous amebic encepha-

Adapted from the author's presentation at the Workshops in
Clinical Lab Sciences Continuing Education Program, "Con-
temporary Protozoal Pathogens: Acanthamoeba, Naegleria,
Leptomyxid Ameba, and Microsporidia," at The Mount Sinai
School of Medicine, New York, NY, on October 25, 1991. Final
revision received March 1992. From the Parasitic Diseases
Branch, Division of Parasitic Diseases, National Center for
Infectious Diseases, Centers for Disease Control, Public
Health Service, U.S. Department of Health and Human Ser-
vices, Atlanta, GA. Address reprint requests to the author at
the Parasitic Diseases Branch, M.S.F./13, Centers for Disease
Control, Atlanta, GA 30333.

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

litis (GAE), a chronic disease that may last from
one week to several months. Cysts of Acan-
thamoeba are usually seen in the brain tissue of
patients with GAE (Fig. 3).

As many as 20 species of Acanthamoeba,
placed in three groups, have been described in the
literature. Distinguishing between members of
groups using morphologic characteristics is rela-
tively easy, but difficulty arises in identifying
members within a group, especially group 2, to
the species level on morphologic grounds alone.
Hence, nonmorphologic criteria such as antige-
nicity, isoenzyme profiles, and restriction length
polymorphism of genomic DNA are being used
increasingly to differentiate the species of Acan-
thamoeba (5—7). Rabbit antisera against 15 of
these species have been prepared at the Centers
for Disease Control (CDC), and it is possible to
identify the species involved, especially in tissue
sections, by the judicious use of these antibodies
in the indirect immunofluorescence test.

In some instances, amebic trophozoites and
cysts in the brain sections of GAE patients have
failed to react with any of the antisera, indicating
that there are probably species of ameba other
than Acanthamoeba and Naegleria that cause hu-
man disease. We at CDC have recently isolated
an unusual ameba from the brain sample of an
encephalitic baboon from the San Diego Zoo Wild
Animal Park (4). Normal procedure for the isola-
tion of small free-living amebas is to inoculate
brain, corneal, or environmental samples onto
nonnutrient agar plates covered with a layer of
bacteria such as Escherichia coli or Enterobacter
aerogenes. We therefore inoculated the samples
ontoE. co/j-coated agar plates. We also inoculated
the samples onto African green monkey kidney
(E6) cells as well as human embryonic lung cells.

283

284

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

Fig. 1. Brain section of patient with primary amebic meningoencephalitis due to N. fowleri. Left Note numerous amebic
trophozoites and extensive destruction of brain tissue (original magnification x250). Right Higher magnification showing
characteristic nuclear morphology of amebas (original magnification x 1,100). Cysts are not seen in sections.

Fig. 2. Immunofluorescence staining of iV. fowleri amebas in
brain section of patient with primary amebic meningoenceph-
alitis (original magnification xllOO). Section was reacted
first with monoclonal antibody IV-Dl-30, subsequently with
goat anti-mouse IgG conjugated with fluorescein isothiocy-
anate.

Fig. .3. Brain section of patient with granulomatous amebic encephalitis. Left Note numerous Acanthamoeba trophozo-
ites (original magnification x250). Right Trophozoites and cysts with characteristic double wall are seen in and around a
blood vessel (original magnification x675).

Vol. 60 No. 4

EPIDEMIOLOGY OF INFECTIONS— VISVESVARA

285

The agar plates were negative for amebas even
after three weeks of incubation. However, in the
monkey kidney cell cultures, unusual-looking
amebic organisms became noticeable after three
weeks of incubation. These structures were mor-
phologically unlike Acanthamoeba and Naegleria
and also did not react with any of the Acan-
thamoeba or Naegleria sera, confirming that the
ameba in question was neither Acanthamoeba
nor Naegleria. We subsequently identified this
agent as leptomyxid ameba, belonging to the or-
der Leptomyxida, previously thought to be a
group of innocuous soil organisms (4). We made
rabbit antisera against this ameba, and the anti-
sera, when used on the Formalin-fixed brain sec-
tions of the baboon, reacted profusely with the
ameba in the tissue sections, confirming that this
was a new infectious agent that causes central
nervous system disease in baboons and possibly
in humans. We then tested the anti-leptomyxid
sera on tissue sections of several previously un-
identified cases of GAE in the indirect immuno-
fluorescence assay. The amebas in these tissue
sections all reacted profusely with the anti-lepto-
myxid serum but not with any of the Acan-
thamoeba or Naegleria sera, confirming the no-
tion that this ameba also infects humans.

Lifecycles of
Free-Living Amebas

N. fowleri is thermophilic and tolerates tem-
peratures of 40°C-45°C (1). Its lifecycle consists of
a trophozoite, a flagellate, and a cyst stage. Under
certain circumstances, such as a change in the
ionic concentration of the environment, the tro-
phozoite of A^. fowleri transforms into a nonfeed-
ing flagellate stage. The flagellate will eventu-
ally revert back to the trophic stage. The
trophozoite measures 10-20 ixm and is character-
ized by the presence of a nucleus with a large
central nucleolus. The trophozoite feeds on bacte-
ria such as E. coli and rapidly multiplies. It also
differentiates into a round, smooth, double-walled
cyst during adverse conditions.

Acanthamoeba is slightly larger and has two
stages in its life cycle, the trophozoite and the
cyst. The trophozoites oi Acanthamoeba may mea-
sure anywhere from 10 to 50 ixm, depending on
the species. They are characterized by a number
of spinelike processes termed acanthopodia that
protrude from the surface of the body. Acan-
thamoeba has no flagellate stage. The trophozoite
differentiates into a characteristically double-
walled cyst during unfavorable conditions. The
outer ectocyst is usually wrinkled, whereas the

PAM GAE GAE

N. fowleri Acanfhiamoeba spp. Leptomyxid

Fig. 4. Reported cases of primary amebic meningoencepha-
litis (PAM) and granulomatous amebic encephalitis (GAE) in
the United States, as of September 1, 1991.

inner endocyst is either stellate, polygonal, oval,
or even round.

Leptomyxid ameba is larger than Acan-
thamoeba and also possesses a cyst that, at the
light microscope level, may be confused with that
oi Acanthamoeba. The trophozoites are irregular
and sometimes highly branched and measure
from 50 to 60 |xm in length. The cysts, which are
irregularly round, measure 15-30 iJim in diame-
ter and appear to have two walls at the light mi-
croscope level. However, at the electron micro-
scope level, three layers of cyst wall may be
discerned: an outer wavy wall, a middle struc-
tureless layer, and an inner thin endocyst (4).
Both the trophozoite and the cyst are uninucleate,
but occasionally a few binucleate forms may be
seen.

The first reported case of amebic meningoen-
cephalitis caused by free-living amebas was de-
scribed from Australia by Fowler and Carter in
1965 (7a). At that time it was thought to be due to
Acanthamoeba, but later the organism was rei-
dentified as N. fowleri based on the epidemiologic
data, clinical picture, and the small size of the
amebas in the tissue sections.

Of the 116 cases of amebic encephalitis re-
corded at CDC from all over the United States,
59% are due to N. fowleri, 29% are due to Acan-
thamoeba spp., and 12% are due to leptomyxid
ameba (Fig. 4).

Primary Amebic Meningoencephalitis. The
cases of primary amebic meningoencephalitis
(PAM) that occurred in 1937 and in the 1950s
were diagnosed retrospectively after the first re-
ported case in 1965. Of the 70 cases of PAM, 71%
have occurred in males and 29% in females. Al-
though patients have ranged in age from 5 years
to 60 years, the majority of cases have occurred in
the 15-18 year age group, primarily in healthy

286

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

young individuals with a history of participating
in water-related sports.

The only patient known to have survived this
disease in the United States was an 8-year-old
girl in 1978 who, along with her parents, had a
history of swimming in a hot spring in Southern
California. The patient was aggressively treated
with amphotericin B, miconazole, and rifampin
(8). She is one of the few patients who developed
antibodies to N. fowleri, with a titer of 4096 in the
indirect immunofluorescence test (2). An immu-
noblot test revealed that the antibody belonged to
the IgM class (2). IgM antibody also was present
in two other patients who died of this disease de-
spite treatment with the same drugs given to the
8-year-old girl (2). N. fowleri amebas isolated
from the cerebrospinal fluid of this girl were ini-
tially considered to be less virulent organisms;
however, mouse pathogenicity tests later re-
vealed this strain to be highly virulent, as it
killed mice within 5 to 7 days (unpublished data).
Isoenzyme and restriction fragment length poly-
morphism of the total DNA analysis of this strain
both confirmed that this strain is a typical N.
fowleri (2, 9).

Granulomatous Amebic Encephalitis. Acan-
thamoeba spp. cause Acanthamoeba keratitis,
Acanthamoeba sinusitis, and of course granulo-
matous amebic encephalitis (GAE) (1-3).

At least 48 cases of GAE have occurred in the
United States. Thirty-four of these cases were due
to Acanthamoeba and the other 14 were due to the
leptomyxid ameba. The portal of entry is not
clearly known; it could be the upper respiratory
tract or skin abscesses. A classic case is that of a
30-year-old Mexican woman who was bitten on
the right arm by her daughter (10). The bite area
became swollen, red, and indurated, and the in-
flammation grew for 6 months, after which time
she was diagnosed to have tuberculosis and was
admitted to a sanatorium in San Francisco. Sev-
eral punch biopsies of the affected area did not
disclose the infectious agent. A few months later
she developed neurologic symptoms, and a brain
biopsy revealed a large number of amebas. All
efforts to culture the organisms were unsuccess-
ful.

As many as 20 cases oi Acanthamoeba infec-
tions have been reported in patients with AIDS.
Since the responsible organism has been isolated
from only a few cases, specific identification is
based on the immunofluorescence reactivities of
the amebas in the brain sections. Skin abscesses
are a common occurrence in AIDS patients with
Acanthamoeba infection (11, 12); Acanthamoeba

organisms have also been isolated from these skin
abscesses.

Of the 14 cases of GAE due to the leptomyxid
ameba, four occurred in patients with AIDS. Of
the 10 cases in non-AIDS patients with the lepto-
myxid ameba, four occurred in patients 60 years
or older, the remaining six in patients under 11
years of age.

Acanthamoeba Keratitis. The first docu-
mented case of Acanthamoeba keratitis ( AK) oc-
curred in 1972 in a Texas farmer who was struck
in his eye by a straw while carrying a bale of hay
(13). He later developed redness of the eye and
was treated for conjunctivitis. A few weeks later
he developed unbearable pain, and a corneal bi-
opsy revealed Acanthamoeba cysts.

From 1972 through 1984 only 12 cases of
keratitis were reported to CDC (14). However, as
many as 24 cases were reported in 1985. Exami-
nation of patients' records revealed that 20 of the
24 cases were in contact-lens wearers (14). A pro-
spective and retrospective epidemiologic study
carried out on 200 AK patients and 600 controls
(matched for geographic location, sex, and age)
revealed that the risk factors for Acanthamoeba
keratitis were exposure to contaminated water,
corneal trauma, contact lens wear, and the use of
unsterile homemade saline solution (15, 16). Sev-
enty-eight percent of patients with AK and 17%
of controls used homemade saline as both a pre-
disinfection wetting solution and a postdisinfec-
tion soaking agent (15). Many patients also had
histories of swimming in lakes, ponds, and pools
while wearing contact lenses. The warning issued
by the Food and Drug Administration and by the
Contact Lens Association of Ophthalmologists on
the use of unsterile homemade saline solutions is
believed to be responsible for the decrease in the
number of Acanthamoeba keratitis cases in re-
cent years.

Laboratory Diagnosis of
Microsporidia Infections

Microsporidia are typical eukaryotic microor-
ganisms, except for their ribosomes, which are
prokaryotic. They are obligate intracellular par-
asites that occur worldwide and cause disease in
almost all major phyla, including protozoans (17).
They produce resistant spores and have a unique
method of infection. Because of their unique char-
acteristics, they are included in a separate phy-
lum, Microspora. A number of genera in the order
Microsporida, including Encephalitozoon, Nose-
ma, Pleistophora, and Enterocytozoon, are known

Vol. 60 No. 4

EPIDEMIOLOGY OF INFECTIONS— VISVESVARA

287

to cause disease in humans (17, 18). The order
Microsporida is divided into two suborders:
Pansporoblastina, in which sporogonic develop-
ment takes place in a sporophorous vesicle, the
pansporoblastic membrane (e.g., Pleistophora),
and Apansporoblastina, in which the sporogonic
development occurs in the cytoplasm in the ab-
sence of a sporophorous vesicle (e.g., Encephalito-
zoon, Nosema, Enterocytozoon).

Enter ocytozoon is the principal cause of diar-
rhea in patients with AIDS. Some of the enteric
problems relate to villus atrophy and fusion,
crypt elongation, mild inflammatory infiltrate,
depletion of goblet cells, and enterocyte vesicula-
tion and swelling (18). Other genera cause kera-
titis, myositis, hepatitis, and peritonitis.

Microsporidia are gram-positive. Brown and
Bren or Brown and Hops modification of Gram
stain can be used with good result. However, elec-
tron microscopy is needed to visualize the inter-
nal structures. The microsporidian spore mea-
sures about 1x2 |jLm and has a complex
structure. The spore has a thin electron-dense
exospore, a thick electron-lucent endospore, and a
thin cell membrane. It has a coiled polar tubule,
which is extruded. The infectious sporoplasm
travels through the polar tubule and is injected
into the host cell. In the transmission electron
micrographs the five to seven spherical units that
appear on either side of the spore are the cross
sections of the coiled polar tubule.

We have recently isolated in culture a mi-
crosporidian parasite from the urine of an AIDS
patient (Fig. 5). It has been continuously grovm in
monkey kidney cells as well as in human lung
fibroblast cell cultures (19). The spores are peri-
odically harvested from the culture flasks by
pouring off the supernatant and centrifuging the
medium. Fresh medium is then added to the cell
culture. The cultures can be maintained and
spores harvested for several weeks. After several
such harvests, when most of the cells are gone,
usually in 8-10 weeks, the medium turns alka-
line and spores are usually seen with their polar
tubules extruded. To initiate fresh cultures, the
sedimented spores are inoculated into a fresh cul-
ture flask, or infected cells from the seed flask are
scraped and inoculated into fresh cultures.

Spores obtained from cultures can be used as
antigens to detect antibodies in the serum of in-
fected patients. Our patient's serum reacted with
the antigens from the culture in the indirect im-
munofluorescence test to a titer of 64. A better
diagnostic test — either a serologic or a DNA-
based test — is needed so that time-consuming and

Fig. 5. Microsporidia spores in urine sample stained with
Gram stain (original magnification xllOO).

cost-prohibitive electron microscopy need not be
used.

Gram stain or Giemsa stain is currently used
to diagnose microsporidia in stool samples (20).
One of the biggest drawbacks of this type of stain-
ing is that everything on the slide, including bac-
teria and yeasts, will stain, making it difficult to
identify the microsporidia. A differential stain
that stains only microsporidia would greatly sim-
plify identification. Dr. Rainer Weber, a visiting
scientist at CDC, has recently developed such a
differential stain that is basically a modification
of trichrome stain (21). The stain uses a tenfold
increase in the concentration of Chromotrope R
over that of the trichrome stain. Microsporidia
spores are stained pinkish red, while bacteria and
yeast are stained green. By allowing for differen-
tial staining of microsporidia spores in complex
specimens such as stool samples, this new stain
will facilitate epidemiologic investigations.

Summary

Acanthamoeba, leptomyxid ameba, and mi-
crosporidia have recently been recognized as or-
ganisms that cause opportunistic infections in pa-
tients with HIV and AIDS. As the numbers of
patients with HIV and AIDS increase, more cases
of amebic encephalitis and microsporidiosis are
likely.

Acknowledgments

The author thanks Sara Wallace for expert technical assis-
tance and Dr. Ralph Bryan for providing data on microspori-
diosis patients.

References

1. Ma P, Visvesvara GS, Martinez AJ, Theodore FH, Dag-
gett P-M, Sawyer TK. Naegleria and Acanthamoeba in-
fections: a review. Rev Infect Dis 1990; 12:490-513.

288

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

2. Visvesvara GS, Stehr-Green JK. Epidemiology of free-liv-

ing ameba infections. J Protozool 1990; 37:25S-33S.

3. Martinez AJ. Free-living amebas: natural history, pre-

vention, pathology and treatment of disease. CRC
Press, Inc., Boca Raton, Florida, 1985.

4. Visvesvara GS, Martinez AJ, Schuster FL, Leitch GJ,

Wallace SV, Sawyer TK, Anderson M. Leptomyxid
ameba, a new agent of amebic meningoencephalitis in
humans and animals. J Clin Microbiol 1991; 28:2750-
2756.

5. De Jonckheere JF. Taxonomy. In: Rondanelli EG, ed. Am-

phizoic amoebae human pathology. Padua, Italy: Piccin
Nuova Libraria, 1987:25^8.

6. Visvesvara GS. Classification of Acanthamoeba. Rev In-

fect Dis 1991; 13(Suppl 5):S369-372.

7. Seidel JS, Harmatz P, Visvesvara GS, Cohen A, Edwards

J, Turner J. Successful treatment of primary amebic
meningoencephalitis. N Engl J Med 1982; 306:346-348.
7a. Fowler M, Carter RF. Acute pyogenic meningitis probably
due to Acanthamoeba spp: a preliminary report. Br Med
J 1965; 2:240-242.

8. Visvesvara GS. Laboratory diagnosis. In: Rondanelli EG,

ed. Amphizoic amoebae human pathology. Padua, Italy:
Piccin Nuova Libraria, 1987:193-215.

9. De Jonckheere JF. Characterization of Naegleria species

by restriction endonuclease digestion of whole-cell
DNA. Mol Biochem Parasitol 1987; 24:55-66.

10. Gullett J, Mills J, Hadley K, Podemski B, Pitts L, Gelber

R. Disseminated granulomatous Acanthamoeba infec-
tion presenting as an unusual skin lesion. Am J Med
1979; 67:891-896.

11. Gonzalez MM, Gould E, Dickinson G, Martinez AJ, Vis-

vesvara G, Cleary TJ, Hensley GT. Acquired immuno-
deficiency syndrome associated with Acanthamoeba in-
fection and other opportunistic organisms. Arch Pathol
Lab Med 1986; 110:749-751.

12. Wiley CA, Safrin RE, Davis CE, Lampert PW, Braude AI,

Martinez AJ, Visvesvara GS. Acanthamoeba meningo-
encephalitis in a patient with AIDS. J Infect Dis 1987;
155:130-133.

13. Jones DB, Visvesvara GS, Robinson NM. Acanthamoeba

polyphaga keratitis and Acanthamoeba uveitis associ-
ated with fatal meningoencephalitis. Trans Ophthal-
mol Soc UK 1975; 95:221-232.

14. Centers for Disease Control. Acanthamoeba keratitis as-

sociated with contact lenses — United States. MMWR
1986; 35:405^09.

15. Stehr-Green JK, Bailey TM, Visvesvara GS. The epide-

miology of Acanthamoeba keratitis in the United
States. Am J Ophthalmol 1987; 107:331-336.

16. Stehr-Green JK, Bailey TM, Brandt FH, Carr JH, Bond

WW, Visvesvara GS. Acanthamoeba keratitis in soft
contact lens wearers. JAMA 1987; 257:57-60.

17. Canning EU, Lorn J, Dykova I. The microsporidia of ver-

tebrates. London: Academic Press, 1986.

18. Bryan RT, Cali A, Owen RL, Spencer HC. Opportunistic

pathogens in patients with AIDS. In: Sun T, ed. Prog-
ress in clinical parasitology, vol 2. Philadelphia: Field
& Wood, 1991:1-26.

19. Visvesvara GS, Leitch GJ, Moura H, Wallace S, Weber R,

Bryan RT. Culture, electron microscopy, and immuno-
blot studies on a microsporidian parasite isolated from
the urine of a patient with AIDS. J Protozool 1991; 38:
105S-111S.

20. Van Gool T, HoUister WS, Schattenkerk WE, et al. Diag-

nosis of Enterocytozoon bieneusi microsporidiosis in
AIDS patients by recovery of spores from faeces. Lancet
1990; 336:697-698.

21. Weber R, Bryan RT, Owen RL, Wilcox CM, Gorelkin L,

Visvesvara GS, and the Enteric Opportunistic Infec-
tions Working Group. N Engl J Med 1992; 326:161-166.

Grand
Rounds

Stroke Prevention in Atrial Fibrillation

Jonathan L. Halperin, M.D., and Elizabeth B. Rothlauf, M.S., R.N.

Abstract

Atrial fibrillation (AF) is a risk factor for ischemic stroke. In randomized trials, AF raised
the risk of stroke nearly sixfold, cumulating in a 35% risk over a lifetime. Anticoagulation
with warfarin reduces the danger of ischemic stroke, but carries hemorrhagic risks, mak-
ing this agent unsuitable for treating many patients. Platelet inhibitor therapy with
aspirin was highly effective for patients younger than 75 years of age in one study, but the
reason for lower efficacy in older individuals is perplexing. These trials support a throm-
botic mechanism for most strokes in patients with AF, but leave physicians in a quandary
as to selection of optimum prophylaxis. Secondary analysis of patients given placebo
identified predictors of thromboembolism, including a history of hypertension, congestive
heart failure, and prior stroke or transient ischemic attack, and echocardiographic find-
ings of left ventricular dysfunction or left atrial enlargement. The absence of these risk
factors selects a fairly large subgroup of AF patients at comparatively low risk of stroke,
for whom the danger and inconvenience of chronic anticoagulation may not be warranted.
It is becoming clear that specific clinical and echocardiographic features allow individu-
alized antithrombotic approaches within the broad category of patients with AF, to en-
hance therapeutic benefit while minimizing hemorrhagic risk.

Effective strategies for prevention of stroke in
patients with atrial fibrillation require first that
the problem of cardiogenic embolism be placed in
perspective. The vast majority of clinical stroke
events — fully 85% — are related to ischemic in-
farction of the brain; the rest are due to hemor-
rhage. Of these ischemic strokes, most are due to
diseases of the extracranial or intracranial arter-
ies; no more than 15% can be specifically traced to
thromboembolism from the cardiac valves or
chambers. At least half the cases of suspected
cardiogenic embolism occur in patients with a
history of atrial fibrillation, but some have
rheumatic heart disease, prosthetic valves, hy-

Adapted from JLH's Grand Rounds in Medicine presentation
on March 10, 1992, at The Mount Sinai Medical Center. Final
revision received February 16, 1993.

From the Division of Cardiology, Department of Medi-
cine, The Mount Sinai Medical Center, New York, New York.

Address reprint requests to Jonathan L. Halperin, M.D.,
Division of Cardiology, Box 1030, Mount Sinai Medical Cen-
ter, Fifth Avenue at 100th Street, New York, NY 10029.

Supported by grant (ROl-NS-24224) from the Division of
Stroke and Trauma, National Institutes of Neurological Dis-
orders and Stroke.

pertension, or left ventricular dysfunction, which
are independently associated with stroke (Fig. 1)
(1). About one in four embolic episodes is related
to ventricular thrombi, as occur in patients with
acute myocardial infarction, particularly within
the first 10 days of large anteroapical infarcts, or
in patients with chronic cardiomyopathy (2). Pa-
tients with left ventricular aneurysm formation
are at lower risk of embolism even though mural
thrombus is frequently present, perhaps because
stasis prevails in the aneurysmal sac so that dy-
namic circulatory forces do not mobilize the
thrombotic mass (3).

Particular thromboembolic risk is associated
with nonvalvular (nonrheumatic) atrial fibrilla-
tion, which raises the chance of stroke about six-
fold, to more than 5% per year. This risk was first
recognized in epidemiologic projects such as the
Framingham Heart Study, which found an even
stronger association of nonvalvular atrial fibril-
lation with stroke among elderly patients (4).
Many of the ischemic strokes that occur in pa-
tients with atrial fibrillation may not have a
cardioembolic mechanism, because coexisting
hypertension or atherosclerosis contributes to ce-

The Mount Sinai Journal of Medicine Vol 60 No. 4 September 1993

289

290

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

Acute Myocardial Chronic
Infarction LV Dysfunction

Fig. 1. Nonvalvular atrial fibrillation is associated with
nearly half the cases of cardiogenic embolism; valvular heart
disease, acute and chronic left ventricular dysfunction, and
other disorders account for the remainder (10).

rebrovascular pathology, and it is seldom certain
whether a given clinical event was caused by em-
bolism of thrombotic material originating in the
fibrillating left atrium. Carotid atherosclerotic
disease may account for some cases, but Wein-
berger et al., found a potential carotid source for
cerebral ischemic symptoms in fewer than a third
of patients with nonvalvular atrial fibrillation ex-
amined by ultrasound imaging of the extracra-
nial arteries (5). Until the results of randomized
trials became available, prophylactic antithrom-
botic therapy was based entirely on a theoretical
concept: atrial fibrillation creates a situation of
stasis within the left atrium, particularly the
atrial appendage, leading to formation of fibrin
and thrombin. Administration of an anticoagu-
lant such as warfarin has been considered appro-
priate in patients at high risk, though risk groups
were poorly defined. Platelet activation has been
thought less important in the pathogenesis of car-
diogenic embolism than in atherosclerotic dis-
ease, so a role for platelet inhibitor medication
such as aspirin was until quite recently entirely
speculative (6).

Refining Definitions of Risk Groups. Popu-
lation-based data indicate that within the broad
category of patients with atrial fibrillation, not
all are at equal risk of stroke. Those at high risk
(event rates over 6% per year) are patients with
rheumatic mitral stenosis, previous cerebral isch-
emia or systemic embolism, prosthetic heart
valve (particularly mechanical prosthesis in the
mitral position), or dilated cardiomyopathy (7).
On the other end of the spectrum are patients
with "lone" atrial fibrillation, younger individu-
als without organic cardiopulmonary disease in
whom atrial fibrillation is often paroxysmal or
intermittent (8). A series of 97 such patients un-
der age 60 years followed over a period of 17 years
had an incidence of ischemic stroke well below 1%
per year, comparable to people in sinus rhythm (9).

Between these poles lies the majority of pa-
tients with nonvalvular atrial fibrillation, who
were, until recently, at uncertain or intermediate
risk. Here, too, lies a great deal of medical lore:
false or unverified teachings based on notions of
pathogenesis that have not been evaluated in
properly conducted randomized trials. These no-
tions are the source of some of the concern about
thromboembolism surrounding restoration of si-
nus rhythm by chemical or electrical cardiover-
sion. Although paroxysmal atrial fibrillation nec-
essarily entails frequent spontaneous conversion
to sinus rhythm, there is no greater risk of stroke
than with constant fibrillation (10). Similarly,
while the Framingham Study suggested an in-
creased risk of stroke during the period immedi-
ately after detection of atrial fibrillation, in re-
cent studies the risk appears cumulative, events
occurring at a more or less constant rate over
years (11). The diagnosis of "silent" stroke, based
on areas of infarction on computerized-tomo-
graphic or magnetic resonance images of the
brain, may also identify a group of patients with
atrial fibrillation at greater risk of clinical stroke;
the same may prove true when there is carotid
atherosclerotic disease, but the importance of
these additional risk factors has yet to be verified
(12).

To the cardiologist, the most important issue
is the underlying etiology of atrial fibrillation,
best defined by clinical history and echocardiog-
raphy. Mitral regurgitation appears to reduce the
risk of stroke by minimizing stasis and thrombo-
genicity within the left atrium. Only recently has
enlargement of the left atrium been validated as
a predictor of stroke risk in patients with nonval-
vular atrial fibrillation, just as it is in cases of
valvular heart disease. Even more important, the
finding of impaired left ventricular function has
been recognized as a powerful marker of throm-
boembolic risk (13). Left atrial thrombi, only
rarely identified by transthoracic echocardiogra-
phy, may not be directly associated with an in-
creased risk of stroke, since the clinical problem
is not one of thrombus formation, but rather of
embolic migration to the vessels supplying the
brain (14).

Who Are the Candidates for Anticoagulation
Therapy? Whatever our notions of pathogenesis,
nonvalvular atrial fibrillation carries a risk of
stroke at least six times that of patients in sinus
rhythm. The scope of the problem is considerable:
over a million North Americans with a 35% life-
time risk of stroke in the absence of antithrom-
botic therapy. These neurologic events are often
devastating, stealing language and movement

Vol. 60 No. 4

STROKE PREVENTION IN AF— HALPERIN AND ROTHLAUF

291

and dehumanizing victims. The toll in human
terms is inestimable, but the cost of acute and
convalescent care for stroke victims comes to
nearly a billion dollars a year in the United
States alone. Although potent antithrombotic
medication has been available for decades, it has
until recently been unclear which patients should
be exposed to the hemorrhagic risks of anticoag-
ulation, particularly as no alternative treatment
had proven prophylactic value (15).

In 1987, investigators sponsored by the Na-
tional Institute of Neurological Disorders and
Stroke — 70 cardiologists, neurologists, and re-
search nurses from 15 medical centers — began
working together to evaluate antithrombotic
therapy for patients with nonvalvular atrial fi-
brillation. The initial Stroke Prevention in Atrial
Fibrillation (SPAF) study was a randomized clin-
ical trial comparing warfarin, aspirin, and pla-
cebo in patients with electrocardiographically
documented atrial fibrillation (sustained or inter-
mittent) without mitral stenosis or prosthetic
heart valves, who did not have either a proven
medical need for or contraindication to an-
tithrombotic therapy.

Patients considered candidates for anticoag-
ulation entered group I of this two-part trial and
were randomly assigned to treatment with war-
farin, aspirin, or placebo. Those with contraindi-
cations to anticoagulation or an unwillingness to
accept the risk and inconvenience of prothrombin
time monitoring necessary with this form of ther-
apy entered a parallel trial, group II, in which
treatment consisted of either aspirin or placebo
(Fig. 2) (16). It was projected that 1644 patients
would be required to compare aspirin, 325 mg
daily in enteric-coated form, and warfarin, dose
adjusted to raise prothrombin time to 1.3-1.8
times control, with placebo for reduction in the
primary end-point events — ischemic stroke and
systemic embolism. Only aspirin and placebo
were given in double-blind fashion. To compen-
sate for the unblindedness of warfarin treatment,
a method of blinded-event verification was em-
ployed.

Aspirin or Warfarin? After the trial had pro-
gressed for little more than a year, a safety-mon-
itoring committee recommended stopping ad-
ministration of placebo because of the clear supe-
riority of active antithrombotic medication for
stroke prevention (Fig. 3). A dramatic reduction
(by more than 80%) in the incidence of ischemic
stroke and systemic embolism was evident in
group I (warfarin-eligible) patients given either
aspirin or warfarin. It became clear that patients
with nonvalvular atrial fibrillation should be

Eligible, Consenting Patients
with Nonvalvular Atrial Fibrillation

n = 1,330

Anticoagulation
Candidates
(Group 1)

n = 627

Non-anticoagulation
Candidates
(Group II)

n = 703

Warfarin Aspirin Placebo
210 206 211

Aspirin
346

Placebo
357

Fig. 2. Design of the Stroke Prevention in Atrial Fibrillation
study. Eligible consenting patients were categorized as war-
farin eligible (group I) or warfarin ineligible (group II) (16).

given some form of antithrombotic medication,
but the study had insufficient statistical power to
determine which type (aspirin or warfarin) was
more effective. Given the small number of isch-
emic events in patients taking these medications,
even had all strokes occurred in patients assigned
to receive one agent or the other, the less effective
therapy would still have been 61% better than no
treatment (17).

The effect of aspirin was not uniform, as pa-
tients in group II did not show the same benefit of
the platelet inhibitor as those in group I. This
differential effectiveness of aspirin forced the in-
vestigators to examine reasons patients entered
group II rather than group I. The most frequent
reason was simply the refusal of otherwise eligi-
ble patients to accept potential assignment to
warfarin therapy, and these patients responded to

0 6 12 18 24

Months After Randomization

Warfarin

or Aspirin 393 301 207 131 57

Placebo 195 145 101 65 26

Fig. 3. Rates of stroke and systemic embolism (primary
events) in patients given active therapy (warfarin or aspirin)
or placebo in group I of the Stroke Prevention in Atrial Fi-
brillation (SPAF) study (17). Number of patients appears at
the bottom.

292

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

stroke risk reduction
%

bleeding rate
%7r

AFASAK

1007

0.5

SPAF

1330

1.5

DMM 1 Mr

\J.O

CAFA

383

2.1

SPINAF

538

1.1

Aggregate

1.2

Fig. 4. Five antithrombotic therapy trials for patients with
nonvalvular atrial fibrillation: Copenhagen Atrial Fibrilla-
tion-Aspirin-Anticoagulation (AFASAK) trial, Stroke Preven-
tion in Atrial Fibrillation (SPAF), Boston Area Anticoagula-
tion Trial for Atrial Fibrillation (BAATAF), Canadian Atrial
Fibrillation Anticoagulation (CAFA), Veterans Administra-
tion Cooperative Stroke Prevention in Nonvalvular Atrial Fi-
brillation (SPINAF). The percent risk reduction and bleeding
rates for patients anticoagulated in each trial are presented
(18-21, 24).

Number of Risk Factors

Fig. 5. Predictors of thromboembolism in the Stroke Preven-
tion in Atrial Fibrillation (SPAF) trial. One or more risk fac-
tors was associated with an increased event rate for patients
with nonvalvular atrial fibrillation given placebo rather than
aspirin or warfarin. Nearly half the enrollees with none of
these risk factors had a low event rate even without an-
tithrombotic therapy (13, 22). HTN, hypertension; CHF, con-
gestive heart failure; TE, thromboembolism; LA, left atrium;
LV, left ventricle.

aspirin as favorably as those in group I. The most
frequent clinical reason for exclusion from the an-
ticoagulant group was age over 75 years, in part
because of an initial protocol stimulation that
these older patients be assigned only to aspirin or
placebo. When the effect of aspirin was examined
on the basis of patient age, effective prophylactic
value was confirmed for those under 75 years old,
but in the older group, event rates were identical
whether patients took aspirin or placebo. Al-
though the effectiveness of warfarin for preven-
tion of ischemic stroke was not linked to age,
bleeding complications, particularly, may have
been age-related. Overall, hemorrhagic events
sufficiently severe to require hospitalization,
transfusion, or surgery occurred in nearly 2% of
patients treated with warfarin, even when pro-

AFASAK

SPAF

Aggregate

1 1 1

1 0 -1

Aspirin Better Aspirin Worse

Fig. 6. Event rates of patients assigned to aspirin in the
Atrial Fibrillation-Aspirin-Anticoagulation (AFASAK) trial
and the Stroke Prevention in Atrial Fibrillation (SPAF) trial
(18, 19). Meta-analysis reveals an overall benefit of aspirin for
stroke prophylaxis.

thrombin time values were frequently monitored

(18) .

Over the last three years, the results of five
randomized trials of antithrombotic therapy in-
volving patients with nonvalvular atrial fibrilla-
tion have become available. These include the
Copenhagen Atrial Fibrillation-Aspirin-Antico-
agulation (AFASAK) trial, which found warfarin
but not aspirin (75 mg daily) significantly effec-
tive; the average age of the patients was 75 years

(19) . In the Boston Area Anticoagulation Trial for
Atrial Fibrillation (BAATAF), patients were fol-
lowed longer than in the other trials; warfarin
was effective even at a low intensity, but the con-
trol group did not include a placebo, and nearly
half the patients acknowledged taking aspirin.
This may be one reason the event rate was lower
than in other series (20). The Canadian Atrial
Fibrillation Anticoagulation (CAFA) study was
terminated before statistically significant results
were obtained, because it was not deemed accept-
able to continue placebo in view of the results in
other trials (21). Data from the Veterans Admin-
istration cooperative Stroke Prevention in Non-
valvular Atrial Fibrillation (SPINAF) trial have
been reported only in preliminary form; warfarin
proved effective for stroke prevention in patients
with nonvalvular atrial fibrillation, but aspirin
was not tested (21a).

Taken in aggregate, these studies convinc-
ingly demonstrate that patients with atrial fibril-
lation who receive no active antithrombotic med-
ication have a sixfold greater risk of ischemic
stroke than comparable patients in sinus rhjrthm,
amplifying the conclusions drawn from epidemi-
ologic studies. The rate of disabling strokes is at
least 2.5% per year, and if one includes transient

Vol. 60 No. 4

STROKE PREVENTION IN AF— HALPERIN AND ROTHLAUF

293

ischemic attacks and clinically silent stroke
events detected by cerebral imaging, the risk is
even greater. Warfarin reduces the risk of stroke
by nearly 70%. Bleeding during anticoagulant
therapy occurred at a rate only slightly over 1%
annually, but participants in these trials were

I carefully selected and followed closely according
to strict research protocols (Fig. 4). It seems
likely that broad application in general clinical
practice, particularly when elderly patients are
included, might be associated with substantially
higher bleeding risk.

Identifying Safe Anticoagulation Candi-
dates at High Risk of Stroke. To clinicians, the
data are already sufficient to clarify that war-
farin is effective medicine for patients with atrial
fibrillation. The question has now become not
whether this form of treatment works, but rather
which patients need it. The SPAF investigators
examined entry characteristics of 568 patients as-
signed to the placebo arms using a stepwise
method of secondary analysis. Overall, three fea-
tures available from the clinical history were as-
sociated with statistically significant increased
relative risk of stroke during follow-up: (a) a his-
tory of hypertension, (b) prior thromboembolism
(stroke or transient ischemic attack more than
two years before enrollment), or (c) congestive
heart failure within three months of entry. Trans-
thoracic M-mode and two-dimensional echocar-
diographic findings were also analyzed, and the
finding of left ventricular dysfunction was the
most important echocardiographic predictor of
stroke risk. Left atrial enlargement beyond 2.5
cm/m^ was also correlated with increased stroke
risk (Fig. 5) (22).

Taken together, these clinical and echocar-
diographic parameters were not only useful in
identifying patients with high risk of stroke, but
42% of the population could be identified as at low
risk — too low to justify the hemorrhagic risks,
nuisance, and expense of chronic anticoagulant
therapy. This is particularly true when one con-
siders that a safer alternative to anticoagulation
exists. A meta-analysis of the SPAF and Copen-
hagen AFASAK studies seems to show that aspi-
rin is effective for stroke prevention in atrial fi-
brillation, at least for some patients (Fig. 6).

I There remains the clinical dilemma: which
type of therapy to offer specific patients. For now,
a logical approach would be to administer war-
farin in patients considered safe anticoagulation
candidates, particularly when the risk of throm-
boembolism is high based on a clinical history of
hypertension, thromboembolism, or congestive
heart failure or when left ventricular dysfunction

or enlargement of the left atrium is evident
echocardiographically. It is also clear that rela-
tively low intensity anticoagulation may be suf-
ficient. Safe anticoagulation requires use of the
International Normalized Ratio of prothrombin
suppression to compensate for variable sensitivi-
ties of thromboplastin reagents used for pro-
thrombin time determinations (23).

Patients may be appropriately treated with
aspirin when clinical and echocardiographic fea-
tures suggest a low thromboembolic risk or an
increased risk of bleeding during warfarin anti-
coagulation. The decision might also be based on
predicted aspirin effectiveness, though this re-
quires a good deal of pathophysiologic specula-
tion. Fortunately, trials are currently in progress
comparing warfarin and aspirin in patients with
atrial fibrillation, including the SPAF-II study,
which is examining this issue in parallel popula-
tions older and younger than 75 years of age.

The future may offer even better alterna-
tives, perhaps combining the best of both worlds
by administering low doses of aspirin with very
low intensity anticoagulation, to enhance thera-
peutic benefits while minimizing hemorrhagic
risks.

At this point, one overriding principle is
clear: virtually all patients with nonvalvular
atrial fibrillation should be given warfarin or as-
pirin for stroke prevention. Atrial fibrillation is
associated with up to 100,000 strokes annually in
the United States at a cost approaching $1 billion
each year. The simple prescription of an aspirin
could prevent 150 strokes a day.

References

1. Halperin JL, Hart RG. Atrial fibrillation without valvu-

lar heart disease. In: Butchart EG, Bodner E, eds.
Thrombosis, embolism and bleeding. London: ICR Pub-
lishers, 1992:l&-30.

2. Stein B, Fuster V, Halperin JL, Chesebro JH. Antithrom-

botic therapy in cardiac disease: an emerging approach
based on pathogenesis and risk. Circulation 1989; 80:
1501-1513.

3. Lapeyre AC, Steele PP, Kazmier FJ, et al. Systemic em-

bolism in chronic left ventricular aneurysm. Incidence
and the rate of anticoagulation. J Am Coll Cardiol
1985; 6:534-538.

4. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation: a

major contributor to stroke in the elderly: the Framing-
ham Study. Arch Intern Med 1987; 147:1561.

5. Weinberger J, Rothlauf EB, Materese E, Halperin JL.

Noninvasive evaluation of the extracranial carotid ar-
teries in patients with cerebrovascular events and
atrial fibrillation. Arch Intern Med 1988; 148:1785-
1788.

6. Dunn M, Alexander J, deSilva R, et al. Antithrombotic

therapy in atrial fibrillation. Chest 1989; 95(Suppl):
118.

294

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

7. Roberts WC, Siegel RJ, McManus BM. Idiopathic dilated

cardiomyopathy: analysis of 152 necropsy patients. Am
J Cardiol 1987; 60:1340-1355.

8. Kopecky SL, Gersh BJ, McGoon MD, et al. The natural

history of lone atrial fibrillation: a population based
study over three decades. N Engl J Med 1987; 317:669-
674.

9. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation: a

major contributor to stroke in the elderly: the Framing-
ham Study. Arch Intern Med 1987; 147:1561.

10. Halperin JL, Petersen P. Thrombosis in the cardiac cham-

bers: atrial fibrillation and left ventricular dysfunction.
In: Fuster V, Verstraete M, eds. Thrombosis in cardio-
vascular diseases. Philadelphia: WB Saunders, 1992:
215-236.

11. Wolf PA, Kannel WB, McGee DL, et al. Duration of atrial

fibrillation and imminence of stroke: the Framingham
Study. Stroke 1983; 14:664.

12. Feinberg WM, Seeger JF, Carmody RF, et al. Epidemio-

logic features of asymptomatic cerebral infarction in
patients with nonvalvular atrial fibrillation. Arch In-
tern Med 1990; 150:2340-2344.

13. Stroke Prevention in Atrial Fibrillation Investigators.

Predictors of thromboembolism in atrial fibrillation:
clinical features of patients at risk. The Stroke Preven-
tion in Atrial Fibrillation Study. Ann Intern Med 1992;
116:1-5.

14. Rosenthal MS, Halperin JL. Thromboembolism in nonval-

vular atrial fibrillation: the answer may be in the ven-
tricle. Int J Cardiol 1992; 37:277-282.

15. Halperin JL, Hart RG. Atrial fibrillation and stroke: new

ideas, persisting dilemmas. Stroke 1988; 19:937-941.

16. Stroke Prevention in Atrial Fibrillation Study Investiga-

tors. Design of a multicenter randomized trial for the
Stroke Prevention in Atrial Fibrillation Study. Stroke
1990; 21:538-543.

17. Stroke Prevention in Atrial Fibrillation Investigators.

Preliminary report of the Stroke Prevention in Atrial
Fibrillation Study. N Engl J Med 1990; 332:863-868.

18. Stroke Prevention in Atrial Fibrillation Investigators.

Stroke Prevention in Atrial Fibrillation Study: final re-
sults. Circulation 1991; 84:527-539.

19. Petersen P, Boysen G, Godtfredsen J, et al. Placebo con-

trolled, randomized trial of warfarin and aspirin for
prevention of thromboembolic complications in atrial
fibrillation: the Copenhagen AFASAK Study. Lancet
1989; 1:175-179.

20. The Boston Area Anticoagulation Trial for Atrial Fibril-

lation Investigators. The effect of low-dose warfarin on
the risk of stroke in patients with nonrheumatic atrial
fibrillation. N Engl J Med 1990; 323:1505-1511.

21. Conolly SJ, Laupacis A, Gent M, et al. Canadian Atrial

Fibrillation Study. J Am Coll Cardiol 1991; 18:349-
355.

21a. Ezckowitz MD, Bridgers SL, James KE, and the SPINAF
Investigators. Interim analysis of VA co-operative
study: Stroke Prevention in Nonrheumatic Atrial Fi-
brillation (SPINAF). Circulation 1991; 84(Sup):A-1791.

22. Stroke Prevention in Atrial Fibrillation Investigators.

Predictors of thromboembolism in atrial fibrillation;
echocardiographic features of patients at risk. Ann In-
tern Med 1992; 116:6-12.

23. Bussey HI, Force RW, Bianco TM, et al. Reliance on pro-

thrombin time ratios causes significant errors in anti-
coagulation therapy. Arch Intern Med 1992; 152:278-
282.

Grand
Rounds

Frontiers in the Treatment of
Ischemic Stroke

Stanley Tuhrim, M.D.
Abstract

Advances in the understanding of the pathophysiology of ischemic neuronal death have
led to the development of new approaches to treating acute stroke. Concurrently techno-
logic advances have permitted application of older approaches in a more sophisticated
manner. These advances herald an era in which rapid, precise evaluation and treatment
of the patient with acute stroke may dramatically alter prognosis. This article reviews
some of these advances.

In the past decade, many new insights into
stroke treatment have altered the approach to pa-
tients with cerebrovascular disease. An evolving
understanding of the pathophysiology of the
acute ischemic process has produced new avenues
for possible intervention based on knowledge of
the intracellular mechanisms underlying isch-
emic neuronal death. This article will review
some of the exciting new approaches to acute isch-
emic stroke therapy.

The Ischemic Penumbra

Experimental models of cerebral ischemia
have demonstrated a continuum of changes when
cerebral blood flow (CBF) is reduced below the
normal resting rate of 50-55 mL/100 g/min, lead-
ing first to neuronal dysfunction and then to cell
death.

Adapted from the author's presentation Grand Rounds in
Medicine on March 17, 1992, at The Mount Sinai Medical
Center. Final revision received March 16, 1993.

From the Department of Neurology, Mount Sinai School
of Medicine (CUNY) New York, New York.

Address reprint requests to Stanley Tuhrim, M.D., De-
partment of Neurology, Box 1137, The Mount Sinai Medical
Center, One Gustave L. Levy Place, New York, New York
10029.

Supported in part by grants from NIH (NINDS), ROl
NS29762 and R29 NS27924.

Normal cerebral perfusion pressure main-
tains membrane potentials, allows appropriate
processing of neurotransmitters, and supports
normal cellular architecture. Below 30 mL/100
g/min there is an increase in extracellular hydro-
gen ion concentration. CBF below 20 mL/100
g/min produces an increase in extracellular potas-
sium ion concentration and in intracellular cal-
cium ion concentration that results in a loss of
electroencephalographic activity and evoked po-
tentials. This indicates dysfunction, but not nec-
essarily irreversible cell damage. The potential
for recovery exists if blood flow is restored to ad-
equate levels. CBF below 10-15 mL/100 g/min re-
sults in membrane depolarization and cell death.

The length of time neurons can be deprived of
adequate blood flow and remain viable is unclear,
but dysfunctional cells may exist in this state for
hours and return to normal if blood flow is re-
stored. In acute ischemic stroke, cerebral tissue
immediately adjacent to an infarcted area may be
deprived of sufficient blood flow to function, yet
remain viable. This region, which may be exten-
sive compared to the irreversibly damaged area,
has been termed the ischemic penumbra (1). Its
existence provides a window of opportunity for
remedial intervention larger than the few min-
utes that was assumed when dysfunction was
equated with cell death, but also indicates that
potential for exacerbating injury if the patient is
treated inappropriately.

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

295

296

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

TABLE

Cellular Therapies

Mechanism

Preventing calcium ion flux
Moderating excitatory

neurotoxins
Reducing cerebral oxygen

demands
Neutralizing free radicals

Protecting cell membranes
Reducing lactate production

Example
Calcium channel blockade

NMDA receptor blockade

Hypothermia, barbiturates
Vitamin E, superoxide

dismutase
GMI ganglioside
Lowering blood sugar

Hypertension in Acute Ischemic Stroke

Controlling hypertension is a major compo-
nent of stroke prevention, but most strokes occur
in patients with a history of hypertension. Treat-
ing elevated blood pressure in the setting of acute
stroke can exacerbate the ischemic injury. In the
normal individual, cerebral autoregulation main-
tains CBF near 55 mL/100 g/min at mean arterial
pressures between 55 and 125 mm Hg. However,
chronically hypertensive patients may maintain
autoregulation only in a higher range, perhaps
only beginning at the upper limit of the range in
normal persons (2). Below this level, CBF would
be dependent on systemic arterial pressure, so
that decreasing the systemic pressure to the "nor-
mal" range could decrease CBF below levels crit-
ical to maintaining adequate perfusion, espe-
cially of the ischemic penumbra. This situation
may be even more dangerous in the setting of
acute stroke because autoregulation is impaired
in areas of ischemic brain, increasing the likeli-
hood that lowering systemic pressure will further
compromise marginally perfused tissue (3). Blood
pressure regulation in acute ischemic stroke is a
pervasive issue, since 84% of stroke patients are
hypertensive on hospital admission (although by
poststroke day 10 only 33% remain hypertensive
without antihypertensive therapy) (4). An under-
standing of the concept of the ischemic penumbra
and cerebral autoregulation indicates the pru-
dence of not lowering blood pressure in the set-
ting of acute ischemic stroke unless other vital
organs (e.g., heart, kidney) are compromised or
blood pressure rises to levels associated with hy-
pertensive encephalopathy (i.e., diastolic pres-
sures greater than 130-140 mm Hg) where
further neurologic damage, presumably from ce-
rebral edema, may ensue.

Cellular Therapies

Potential therapies for acute ischemic stroke
can be broadly categorized into one of two ap-

proaches. Cellular therapies, based on an as yet
imperfect understanding of the biochemical con-
sequences of ischemia, attempt to intervene at
the neuronal level. Some of these are listed in the
Table. Reperfusion therapies aimed at restoring
blood flow to the ischemic region are discussed
below.

Early attempts to treat the biochemical con-
sequences of ischemia tried to reduce metabolic
demand. However, anesthetics, barbiturates, and
hypothermia, although seemingly effective in ex-
perimental models of ischemia, have not proven
efficacious in acute ischemic stroke in humans.
Attention has shifted to the role of calcium ho-
meostasis and its disregulation in ischemia. Nor-
mally, extracellular free calcium concentration
are lO'* higher than intraneuronal levels. Main-
tenance of this gradient is crucial to the function
of the neuron. Intracellular calcium homeostasis
is mediated via two types of calcium channels,
voltage-dependent channels (VDC) and receptor
on neurotransmitter-operated channels (ROC),
and may be affected by nonspecific membrane
leakage as well.

Voltage-Dependent Calcium Channels. Cell
membrane depolarization activates VDC, allow-
ing calcium ion influx. Three types of VDC, tran-
sient, intermediate, and long-lasting (or L-type)
channels, have been identified. The L-type chan-
nel is the most important mediator of calcium ion
influx in ischemia and has been the target for
pharmacologic manipulation with at least four
distinct classes of calcium antagonists, of which
the dihydropyridines, including nimodipine and
nicardipine, have been most extensively tested.
By binding to specific recognition sites, these
agents produce a conformational change that re-
sults in inactivation of the L-type channel. In part
because they are lipophilic and can cross the
blood-brain barrier, these agents are relatively
"cerebroselective," exerting their greatest effect
in the central nervous system, increasing cere-
bral blood flow by dilating cerebral vessels via
their effect on smooth muscle cells and blocking
calcium influx into ischemic neurons.

Nimodipine has been studied in at least
seven clinical trials and has consistently been
shown to mitigate the ischemic neuronal damage
secondary to vasospasm following subarachnoid
hemorrhage. However, no effect of nimodipine on
angiographically demonstrable vasospasm has
been found (5).

In effect, treating patients with subarachnoid
hemorrhage with nimodipine is pretreating pa-
tients at high risk for cerebral infarction before
they suffer an ischemic insult. The results have

Vol. 60 No. 4

FRONTIERS IN TREATMENT OF ISCHEMIC STROKE— TUHRIM

297

not been as encouraging when nimodipine has
been tested in ischemic stroke patients treated up
to 48 hr after stroke onset. An initial study dem-
onstrated significantly reduced mortality at one
month in ischemic stroke patients treated with
oral nimodipine, 120 mg daily, compared with
placebo (6), but three subsequent studies failed to
confirm this (7). If only patients treated within 12
hr of onset are considered, however, a beneficial
effect can be noted. Therefore, a trial in which
patients are treated within 12 hr of onset has
been suggested.

Receptor or Neurotransmitter-Operated Cal-
cium Channels. The primary ROC for calcium en-
try is the A''-methyl D-aspartate (NMDA) receptor
channel. Glutamate binds directly to this recep-
tor, allowing calcium influx. Drugs can block this
activity, by competing with glutamate for its
binding site or by binding at another receptor on
the channel, termed the phencyclidine (PCP) re-
ceptor, and directly preventing calcium influx.
This second, noncompetitive, inhibition with
drugs such as MK801 or dextromethorphan has
been shown to reduce infarction size in laboratory
models (8). However, because of undesirable side
effects, including hallucinations and respiratory
depression, associated with these agents, further
refinement may be required before this approach
proves useful. Other means of influencing cal-
cium ion homeostasis, including intracellular cal-
cium chelation and moderating non-NMDA-me-
diated receptor-operated channels, are under
laboratory investigation.

Free radicals containing impaired electrons
can be generated in areas of incomplete ischemia
such as the ischemic penumbra. They may cause
further injury to cells via peroxidation, particu-
larly during reperfusion of an ischemic area.
Drugs that interfere with lipid peroxidation, such
as superoxide dismutase, have been shown to re-
duce infarct size in laboratory models (9). In par-
ticular 21-aminosteroids, a class of compounds de-
rived from methylprednisolone, but lacking both
glucocorticoid and mineralocorticoid effects, are
currently under study in human clinical trials for
both head trauma and ischemic stroke. Another
antioxidant, vitamin E, could also prove useful in
neutralizing the effects of free radicals.

Reperfusion Therapies

Arterial occlusion by thrombosis is the prox-
imate cause of brain infarction. Clot lysis or pre-
vention of clot propagation so as to restore blood
flow or preserve collateral flow would seem an
obvious intervention strategy, yet no such ther-

apy has been proven effective. Several approaches
are, however, currently under investigation.

For 40 years the antithrombotic therapy
most widely favored has been anticoagulation
with heparin, but its use remains controversial.
Results of studies performed to date have been
conflicting, some early studies suggesting a ben-
efit in progressing stroke (10), but more recent
ones failing to show a benefit (11). However, all
studies to date have had significant methodologic
limitations, such as lack of randomization or in-
sufficient sample size. A greater consensus exists
regarding the usefulness of heparin in preventing
recurrent stroke in patients suffering a cardioem-
bolic stroke. A randomized trial of anticoagula-
tion in patients with acute stroke presumed to be
cardioembolic in origin was terminated after only
45 patients had been entered, because anticoagu-
lation was thought beneficial, although even this
conclusion has been questioned (12).

Low-molecular-weight heparinoids have
been studied as a potentially safer alternative to
heparin because they have a negligible effect on
platelets while maintaining potent antithrom-
botic effects and are therefore less likely to
produce bleeding complications or thrombocy-
topenia. A semisynthetic, low-molecular-weight
heparinoid, ORG 10172, is currently under study
in a double-blind, placebo-controlled trial in the
prevention of ischemic stroke progression. This
well-designed study should avoid the method-
ologic limitations of previous ones. In preliminary
studies, this compound has proven relatively safe
when given to acute stroke patients, although 2 of
a total of 83 patients treated in two studies did
suffer fatal hemorrhagic conversions of large in-
farctions (13).

Because of recent successes in treating myo-
cardial infarction, there is renewed interest in
thrombolytic therapy for acute stroke. The endog-
enous thrombolytic system provides a counterbal-
ance to clotting mechanisms in order to maintain
homoeostasis. Formation of thrombus activates
the thrombolytic system, allowing for restriction
and lysis of intravascular thrombosis. Enhancing
this response promotes recanalization of vessels
occluded by thrombus. Tissue plasminogen acti-
vator (tPA) produced by vascular endothelial cells
and urokinase are endogenous activators of the
thrombolytic system; streptokinase is an exoge-
nous substance that can be administered with
similar effect.

Initial studies of thrombolysis following isch-
emic stroke, performed 30 years ago, were dis-
couraging, leading investigators to temporarily
abandon this approach. However, these early at-

298

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

tempts had significant methodological shortcom-
ings, including the lack of availability of com-
puted-tomography (CT) scanning, limiting both
the initial distinction between intracerebral hem-
orrhage and ischemic stroke and the early detec-
tion of iatrogenic intracerebral hemorrhage. Pa-
tients were sometimes treated days after onset,
when restoration of blood flow would be of limited
value. More recently, with the availability of CT
scanning, magnetic resonance imaging, and su-
perselective cerebral angiography allowing deliv-
ery of thromboljrtic agents directly to the throm-
bosed vessel, several reports have indicated
encouraging results (14, 15). Recanalization of
the affected vessel is usually achieved, but since
none of the intra-arterial studies of urokinase or
streptokinase have been randomized or blinded, a
true test of efficacy has yet to be performed.

Studies of intravenous thrombolytic therapy
have generally employed a form of tPA. Two ran-
domized trials of efficacy are currently underway,
but earlier studies intended to test safety and de-
termine appropriate dosing have demonstrated
that arterial recanalization can be achieved. The
rate of hemorrhagic complications reported has
varied widely from 4% to 50%, with a mortality
rate approaching 50% (16).

Although definitive efficacy is lacking, acute
reperfusion therapies show great promise and are
intuitively attractive because they attack the
proximate cause of ischemic stroke. Many ques-
tions remain regarding this approach relating to
timing, method (intravenous or intra-arterial),
dose, agent, and use of concomitant anticoagula-
tion or antiplatelet therapy. Nevertheless, resto-
ration of adequate blood flow appears key to sal-
vaging ischemic brain.

Agents that act at the cellular level have
been demonstrated to protect the ischemic brain
from irreversible damage, partially mitigating
the effects of inadequate blood flow. As-yet-un-
tried combinations of these agents may prove far
more effective than isolated approaches. Poten-
tially, these agents, alone or in combination, may

be able to forstall or limit cell death until ade-
quate blood flow is restored. Ultimately the com-
bination of cellular and reperfusion therapies
may dramatically alter the prognosis of acute
ischemic stroke.

References

1. Astrup J, Siesjo BR, Symon L. Thresholds in cerebral isch-

emia. The ischemic penumbra. Stroke 1981; 12:723-
725.

2. Strandgaard S, Olesen J, Shinkoj E, et al. Autoregulation

of brain circulation in severe arterial hypertension. Br
Med J 1973; 1:507-510.

3. Fieschi C, Angoli A, Battistini N, et al. Derangement of

regional blood flow and of its regulatory mechanisms
in acute cerebrovascular lesions. Neurology 1968;18:
1166-1179.

4. Wallace JD, Levy LL. Blood pressure after stroke. JAMA

1981;2177-2180.

5. Wong MCW, Haley EC. Calcium antagonists: stroke ther-

apy coming of age. Stroke 1990;21:494-501.

6. Gelmers HS, Gortes K, de Weerdt CJ, Wiezer HJA. A

controlled trial of nimodipine in acute ischemic stroke.
N Engl J Med 1988; 318:203-207.

7. Trust Study Group. Randomized, double-blind, placebo-

controlled trial of nimodipine in acute stroke. Lancet
1990; 336:1205-1209.

8. Simon R, Shiraishi K. NMDA antagonist reduces stroke

size and regional glucose metabolism. Ann Neurol
1990; 27:606-611.

9. Schmidley JW. Free radicals in central nervous system

ischemia. Stroke 1990; 21:1086-1090.

10. Carter AB. Anticoagulant treatment in progressing

stroke. Br Med J 1961; 2:70-73.

11. Haley EC, Kassell NF, Tomer JC. Failure of heparin to

prevent progression in progressive ischemic infarction.
Stroke 1988; 19:10-14.

12. Cerebral Embolism Study Group. Immediate anticoagu-

lation of embolic stroke: a randomized trial. Stroke
1983; 14:668-676.

13. Biller J. Medical management of acute cerebral ischemia.

Neurol Clin 1992; 10:63-85.

14. Hacke W, Zeumer H, Ferbert A, et al. Intra-arterial

thrombolytic therapy improves outcome in patients
with acute vertebrobasilar occlusive disease. Stroke
1988; 19:1216-1222.

15. Mori E, Tabuchi M, Yoshida T, Yamadori A. Intracarotid

urokinase with thromboembolic occlusion of the middle
cerebral artery. Stroke 1988; 19:802-812.

16. Brott TG. Thrombolytic therapy for stroke. Cerebrovasc

Brain Metab Rev 1991; 3:91-113.

Urologic Aspects of Prevention of
End-Stage Renal Disease in Spinal

Cord Injury

Jose R. Sotolongo, Jr., M.D.

Years ago the primary cause of death in persons
with spinal cord injury was end-stage renal fail-
ure. The last 10 years have witnessed a revolu-
tion in the way we have come to understand the
physiology of the bladder and how to best manage
its functions after spinal cord damage. As a re-
sult, end-stage renal failure is no longer the lead-
ing cause of death, which is now cardiopulmonary
disorders. A 25-year-old patient with spinal cord
injury has a life expectancy of 50-60 years — dra-
matic testimony to the importance of proper blad-
der management and to the inapplicability of ear-
lier modalities of urologic treatment; it is no
longer acceptable to simply insert a Foley cathe-
ter or a suprapubic tube and assume that the uri-
nary tract has been adequately managed.

In the management of the neurogenic blad-
der in such patients, the primary objective is to
prevent renal damage and subsequent failure.
Continence and the prevention of pressure sores
are still important, but not difficult to manage.

Effects of Spinal Cord Injury on the Blad-
der. In spinal cord injury, there is an interruption
of neurons between the pontine micturition cen-
ter, which coordinates the activities of the blad-
der and external sphincter, and the pudendal and
pelvic nerve nuclei in the sacral cord. The bladder
will therefore contract spontaneously with simul-
taneous contraction of the external sphincter

Adapted from the author's presentation at the conference
"Spinal Cord Injury: Aspects of Specialized Care" held at the
Bronx Veterans Affairs Medical Center on April 22, 1991;
final manuscript received January 1993. From the Spinal
Cord Injury Service, Bronx VAMC, Bronx. Address reprint
requests to the author, 40 Hurley Avenue, Kingston, NY
12401.

(dyssynergia), an undesirable event of grave po-
tential harm in the upper urinary tract.

In addition, the hypogastric nerve arising
from T-11 through L-1 exerts an accommodating
effect, maintaining low pressure intravesically
during the filling phase. This effect is dampened
during the voiding phase, when the pressure in-
creases under the influence of the pelvic nerve.

Receptors and Drug Choices. The alpha ago-
nist receptors are primarily at the bladder neck
and the urethra. If the patient is treated with an
alpha agonist like pseudoephedrine, for example,
the bladder neck and urethra tone will increase.
In fact, this kind of medication is sometimes used
in women with incontinence associated with
coughing or sneezing. Alpha blockers are used to
produce the opposite effect — to relax muscle tone.
Minipress (Prazosin) will relax the bladder neck
area to some extent but will not affect the exter-
nal sphincter, which is unaffected by any oral
agent.

Beta agonists, which act primarily on the
dome of the bladder for relaxation, are not used
clinically to any extent. Instead, anticholinergic
drugs are used to act on the cholinergic receptors
of the pelvic nerve. Oxybutynin (Ditropan) and
propantheline bromide (Probanthine) are the
most commonly used anticholinergic medications.
They block the parasympathetic influence on the
detrusor muscle so that its contractility is im-
paired, sometimes to the point of paralysis. A
cholinergic agonist, such as bethanechol (Ure-
choline), will, conversely, increase the tone of the
bladder.

Clinical Course of Bladder Effects. The ini-
tial effect is a period of inactivity (areflexia) last-
ing from 1 to 24 weeks. Generally within 4 to 6

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

299

300

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

weeks the bladder gradually resumes automatic
contractions but without the influence of the cor-
tex or the pontine micturition center. There is a
gradual return of contractions and the establish-
ment of a permanent contractile pattern as some
of the edema resolves in the cord after the injury.
There may be a resolution of the dyssynergia, and
a more specific pattern of bladder pressures dur-
ing the filling and voiding phases will be estab-
lished. The permanent pattern of injuries low on
the spinal cord, at the conus medullaris or the
Cauda equina, is usually areflexia.

Injuries above the T-6 level have an impact
on the control of sympathetic outflow. This may
result in autonomic dysreflexia, an uncontrolled
discharge from the sympathetic portion of the
thoracolumbar cord. It is triggered by noxious
stimuli arising primarily in the sacral cord der-
matomes. One of the most common stimuli is a
full bladder. Autonomic dysreflexia can lead to
cerebral hemorrhage, since systolic blood pres-
sure can rise rapidly to more than 200 mm Hg.

Dyssynergia often results in high-pressure
voiding, with consequent hydronephrosis and
stone formation. The post void residual (how
much urine is left in the bladder after voiding) is
not as important as the pressure required for the
bladder to expel urine. High pressure in spinal
cord injury occurs if the sphincter does not open at
the time of the voiding phase.

In addition to high-pressure voiding, one may
encounter high-pressure filling, which can be just
as dangerous to renal function if low-pressure ac-
commodation is impaired by injury to the hypo-
gastric nerve or to the cord at the T-11 to L-1
levels.

A urodynamic study is the only way to deter-
mine whether there is high pressure in the blad-
der. A cystometrogram with fluid contrast, and
simultaneous urethral or sphincter pressure mea-
sured during the filling and voiding stages, is
ideal. A 10 French triple-lumen catheter is usu-
ally used to allow measurements of both the blad-
der and the urethral pressure. Fluoroscopy is
needed to monitor the appearance of the bladder
and the bladder neck and to ascertain the pres-
ence or absence of reflux.

Management. When the bladder becomes se-
verely trabeculated (often attaining the appear-
ance of a Christmas tree), continuing high pres-
sure will lead to deterioration of the upper tracts
and eventual end-stage renal failure. There are
two choices in this clinical picture. Intermittent
catheterization can be performed often enough to
prevent the high pressure from occurring. How-
ever, if the frequency of the pressure elevations

noted in the urodynamic tracing indicates that
catheterization would have to be performed more
often than every three hours, this treatment be-
comes impractical. The alternative is to perform a
sphincterotomy by cutting the external sphincter,
thereby allowing the bladder to empty against
low pressure in the urethra.

If the patient is able to perform intermittent
catheterization and prefers not to wear an exter-
nal condom catheter, bladder paralysis can be at-
tempted with anticholinergic medications. The
most frequently used is oxybutynin, a smooth-
muscle relaxant, usually at a dose of 5 mg orally
three or four times a day. It acts as an intracel-
lular calcium channel blocker. Another medica-
tion used frequently is propantheline bromide in
doses of 15 mg orally three times a day. Proban-
theline is primarily a competitive antagonist at
the cholinergic receptor. Together these two
agents have a synergistic effect, since each works
at different sites. Other anticholinergic agents
have no advantage over these two. Tofranil, an
imipramine, used primarily in psychiatry, is oc-
casionally fairly effective in its anticholinergic ef-
fect on the bladder.

It is essential that vesical pressure be moni-
tored in patients on intermittent catheterization.
If the urethral pressure is high, for example 90
cm H2O, and the bladder pressure reaches 45 or
50 cm, the patient will not leak between cathe-
terizations because the resulting pressure in the
bladder is below that of the urethra. However,
this is not a "safe" bladder, because the ureter
cannot propel urine against a resting pressure
above 35 cm H2O. Hydronephrosis will result any
time the resting pressure exceeds 35 cm. It is
therefore important to confirm that low intraves-
ical pressure has been attained with medication,
when necessary, when intermittent catheteriza-
tion is the treatment choice.

The common adverse effects of anticholin-
ergic drugs are dry mouth, constipation, and uri-
nary retention. In spinal-cord-injured patients,
urinary retention may be the desired effect. Vi-
sual disturbances are an indication for discontin-
uation. Anticholinergics should not be prescribed
for patients with glaucoma. These agents should
also be avoided in patients with the Shy-Drager
syndrome because the episodes of hypotension
may be aggravated.

Finally, some patients already have upper-
tract damage as a result of their bladder manage-
ment as practiced 20-30 years ago. Struvite stone
formation is the rule. Extracorporeal lithotripsy
is one of the most important tools available at this
point to avoid end-stage renal failure. Once kid-

Vol. 60 No. 4

UROLOGIC ASPECTS— SOTOLONGO

301

ney stones occur as a result of the high pressure
transmitted to the kidney and the chronic infec-
tion, the course is one of recurring infections and
stone formation.

However, extracorporeal shock wave litho-
tripsy has been a true lifesaver. Previously, such
patients required surgical removal of the stone.
However, stones and infection would recur, re-
quiring second and third surgical interventions,
often resulting in decrease or loss of renal func-
tion. These patients came to dialysis because the
disease was usually bilateral.

Lithotripsy has forever revolutionized that.
We can now treat multiple recurrences. Although
it is not a magic cure, since renal function may
still deteriorate as a result of chronic infection
and stone formation, end-stage renal failure can
be delayed.

Questions and Answers

Question: How successful is transurethral
sphincterotomy?

Sotolongo: It has a 25%-30% failure rate. The
skeletal muscle that is incised may heal readily.
Often the sphincterotomy has to be repeated in
two or three months, sometimes a few years later.
Question: Does an indwelling catheter prevent
high bladder pressure?

Sotolongo: It does prevent high pressure, but at
high cost. The indwelling catheter, either ure-
thral or suprapubic, entails three major compli-
cations. One is bladder stone formation, and a re-
sultant high propensity to form stones in the
upper tract as well. Bladder stones are a bad
omen for these patients. The non-life-threatening
complications are urethral diverticula and ab-
scess and loss of a testicle as a result of epididy-
mitis. Occasionally, a bladder contracts so com-

pletely that it is virtually collapsed. It thus
obstructs the upper tract mechanically. Although
an indwelling catheter does avoid high bladder
pressure, it is not a desirable solution.
Question: What about the new techniques of con-
trolling bladder function electronically?
Sotolongo: I think the question refers to bladder
"pacemakers," which are being studied in Califor-
nia and are becoming somewhat more wide-
spread, although they are still investigational.
The procedure involves placing electrodes in the
sacral cord and actually stimulating certain
nerves that control the external sphincter and
bladder. One or the other can be stimulated, so, in
theory, one should be able to control bladder and
sphincter perfectly. In truth, bladder and sphinc-
ter function rely on a complex neurologic system.
Stimulating only one nerve can cause other re-
flexes that may overcome what one is trying to
accomplish. Therefore, certain nerve roots have to
be cut, resulting in other complications. So, yes, I
think the procedure holds future hope, but I do
not think it is part of standard therapy at this
point, although it may become an effective way of
controlling the bladder more physiologically.
Question: Should I VPs be performed yearly on
spinal-cord-injured patients?
Sotolongo: If we have studied the patient with
urodynamics and we know that the bladder pres-
sure is safe, there is little indication to do an IVP
yearly because we have been preventing the ini-
tiating insult. One can perform yearly renal ul-
trasound studies, which are sensitive in detecting
hydronephrosis as well as stones. If the renal
sonogram is abnormal, we can then go on to the
IVP the same year. It is relatively rare, however,
that the IVP is abnormal if there is documented
low pressure in the bladder.

Pathophysiology of End-Stage Renal
Disease in Spinal Cord Injury

N. D. Vaziri, M.D., FACP

Spinal-cord-injured patients are at increased
risk of developing end-stage renal disease. In fact,
prior to the 1960s, chronic renal failure was the
principal cause of death in as many as 75% of all
spinal-cord-injured (SCI) patients. Fortunately,
during the past two to three decades, considerable
progress has been made in the preservation of re-
nal function and prevention of end-stage renal
disease in this population. Progress is clearly re-
flected in subsequent surveys, which have re-
vealed a steady decline in mortality from renal
failure in SCI patients. For instance, the mortal-
ity from chronic renal failure among World War
II and Korean War veterans, reported in 1974,
approximated 45%, whereas that reported for the
veterans of the Vietnam conflict was 20% (1, 2).
An even lower incidence of about 14% was subse-
quently demonstrated in a large series of civilian
cases followed up by Price (3).

The pathogenesis of chronic renal failure as-
sociated with spinal cord injury is complex and
multifactorial. Impaired urine flow associated
with neuropathic bladder dysfunction plays the
central role in the vicious cycle that can lead to
chronic pyelonephritis and reflux nephropathy,
urolithiasis, and destruction of renal paren-
chyma.

In a large series of patients with end-stage
renal disease due to spinal cord injury studied by
our group, all patients without exception had

Adapted from the author's presentation at the conference
"Spinal Cord Injury: Aspects of Specialized Care" held at the
Bronx Veterans Affairs Medical Center on April 22, 1991;
final manuscript received February 1992. From the Division
of Nephrology, Department of Medicine, University of Cali-
fornia, Irvine. Address reprint requests to the author, who is
Professor and Vice Chairman, Department of Medicine, and
Chief, Division of Nephrology, at the Division of Nephrology,
UCI Medical Center, 101 The City Drive, Orange, CA 92668.

302

chronic pyelonephritis and chronic active urinary
tract infection (4—6). A great majority of these
patients had been maintained on indwelling ure-
thral catheters, some on suprapubic catheters; a
few underwent surgical diversion. Nephrolithia-
sis was extremely common, and renal amyloidosis
was quite frequent. In addition, reflux nephropa-
thy and hypertensive nephrosclerosis were pre-
sent in some cases.

Clearly, impaired urinary drainage predis-
poses to and complicates the treatment of urinary
tract infection, which is both facilitated and per-
petuated by instrumentation, especially indwell-
ing catheters and subsequent development of
nephrourolithiasis. The combination of urinary
tract infection and increased intravesical pres-
sure leads invariably to reflux, which further pro-
motes the destruction of renal tissue. The gradual
reduction in nephron mass due to the combina-
tion of these factors leads to compensatory hyper-
filtration and glomerular capillary hypertension,
which, in turn, accelerates the rate at which re-
maining nephrons are lost. This process is
marked histologically by initial glomerular hy-
pertrophy, followed by glomerular sclerosis.

Thus, when either untreated or improperly
managed, impaired urinary drainage triggers a
chain of interrelated events that can result in
end-stage renal disease. In addition to the pro-
cesses described above, renal amyloidosis is also
commonly present and further contributes to re-
nal deterioration in these patients. The amyloid-
osis in this condition is generally of the secondary
type associated with chronic infections involving
bedsores, osteomyelitis, and possibly chronic ac-
tive urinary tract infection. Accordingly, preven-
tion and treatment of various infections can pre-
clude the occurrence of secondary amyloidosis,
which affects not only the kidneys, but most other
organs as well.

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

Vol. 60 No. 4

PATHOPHYSIOLOGY— VAZIRI

303

Based on this review, it is quite clear that
SCI-associated chronic renal disease is poten-
tially preventable. The most important step in
prevention of renal disease is provision of safe
and satisfactory urine drainage while avoiding
indwelling catheters and diversion techniques
when possible. This is best achieved by promoting
spontaneous voiding or, if that is not possible,
regular clean intermittent catheterization. The
other equally important and related step is pre-
vention and treatment of infection, which directly
(chronic pyelonephritis) and indirectly (struvite
stone, amyloidosis) promotes destruction of renal
tissue. In addition, measures to reduce high in-
travesical pressures associated with spastic blad-
der have been advocated by a number of author-
ities in an attempt to prevent renal tissue damage
by vesicoureteral reflux. Struvite urolithiasis,
which is caused by the presence of urinary tract
infection with urease-producing organisms, itself
can perpetuate the infection, cause obstruction,
and promote kidney damage. Therefore, every ef-
fort should be made to prevent stone disease or
vigorously treat it once it has occurred.

Factors Contributing to Chronic
Renal Disease in SCI Patients

Factors that contribute to chronic renal dis-
ease in spinal-cord-injured patients include:

• Impaired urine flow due to neuropathic bladder
dysfunction

• Urinary infection leading to chronic pyelone-
phritis

• Nephrolithiasis, infection stones, and other cal-
culi

• Reflux nephropathy due to high intravesical
pressure and infection

• Progressive loss of remaining nephrons due to
glomerular hyperfiltration and hypertension

• Secondary amyloidosis

• Hypertensive arteriolonephrosclerosis (uncon-
trolled HTN)

Prevention and treatment of infected or unin-
fected pressure ulcers and osteomyelitis is also
important in preventing damage to the kidney
and other organs from secondary amyloidosis.

In addition to the listed measures of specific
relevance to SCI patients, certain general princi-
ples should be considered. For instance, excessive
use of analgesics can lead to papillary necrosis
and chronic interstitial nephritis and should be
avoided when possible. Similarly, caution should
be exercised in the use of nephrotoxic agents such
as aminoglycoside antibiotics, and control of hy-

pertension, when present, is important to avoid
arteriolonephrosclerosis. In this regard, certain
agents such as angiotensin-converting-enzyme
inhibitors and calcium channel blockers may not
only serve to control hypertension, but also to re-
tard the progression of renal insufficiency when
present.

Measures to Prevent or Retard
Progression of Chronic Renal
Failure in SCI Patients

To prevent or retard progression of chronic
renal failure in patients with spinal cord injury,
the following measures should be taken:

• Provision of safe and effective urine flow

• Prevention or effective treatment of urinary in-
fection

• Correction of high intravesical pressure

• Prevention or treatment of urolithiasis

• Prevention or treatment of pressure ulcers

• Prevention or treatment of chronic extrarenal
infections such as pressure ulcers, osteomyeli-
tis

• Control of arterial hypertension

• Avoidance of excessive or prolonged use of an-
algesics capable of producing analgesic ne-
phropathy

• Caution with the use of nephrotoxic drugs such
as aminoglycoside antibiotics

Interpreting Measures of Renal Function.

Several important issues need to be considered in
interpretation of certain laboratory or derivative
data employed to assess renal function in SCI pa-
tients (6, 7). For instance, serum creatinine con-
centration is generally lower in SCI patients than
in able-bodied individuals with the same glomer-
ular filtration rate. This is due to lower creatinine
production reflecting diminished muscle mass in
SCI patients. Therefore, serum creatinine values
well within the normal range for able-bodied per-
sons may, in a spinal-cord-injured individual, sig-
nify a significant reduction in glomerular filtra-
tion rate. The extent of error is proportional to the
extent of the associated paralysis and muscle at-
rophy, being greater in quadriplegics than para-
plegic persons. In a recent study, Kaji et al.
showed urinary creatinine excretion and creati-
nine clearance to be markedly lower in a group of
spinal-cord-injured patients, serum creatinine
ranging between 1.0 and 0.5 mg/dL when com-
pared with a group of able-bodied individuals (8),
confirming previous observations. Mathematical
formulas devised to predict creatinine clearance
in able-bodied individuals using serum creati-

304

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

nine, age, weight, and sex are highly unreliable
when employed without modification for SCI pop-
ulations. We and others have developed correc-
tion factors to adapt these formulas to paraplegic
and quadriplegic persons (9). Given its limita-
tions, creatinine clearance is a relatively reliable
tool for assessing the renal excretory function;
however, certain precautions have to be taken to
avoid inaccuracies. Clearly, the accuracy of the
result is dependent on the precision and complete-
ness of the timed urine collection in all subjects.
However, Kaw et al. recently reported a signifi-
cant fall in urine creatinine concentration from
the time of collection to the point of assay in SCI
patients with certain urinary infections (10). This
particular phenomenon compounds the complex-
ity of this test and interpretation of its results.

With regard to other laboratory measure-
ments, we have found certain differences between
SCI and able-bodied patients with end-stage re-
nal disease. As a group, the SCI patients had sig-
nificantly lower hematocrit and serum bicarbon-
ate and albumin concentrations. In contrast, they
had higher urine output and proteinuria than
their able-bodied counterparts. Many SCI pa-
tients had nephrotic-range proteinuria while on
maintenance dialysis. This is a relatively un-
usual phenomenon, since glomerular leakage of
protein usually declines in parallel with the de-
cline in nephron mass as reflected by the fall in
glomerular filtration rate. However, renal amy-
loidosis represents an exception to this rule and
accounted for persistent heavy proteinuria in our
patients.

Avoiding Hyperkalemia. Hyperkalemia is a
major and potentially life-threatening complica-
tion of renal insufficiency. Usually progressive
loss of nephron mass associated with chronic re-
nal insufficiency triggers a number of adaptive
processes to mitigate the rise in extracellular po-
tassium concentration. These include readjust-
ments in glomerulotubular balance in the rem-
nant nephrons, leading to an increase in
potassium excretion. In addition, the large intes-
tine assumes a significant role in potassium ho-
meostasis by secreting large quantities of potas-
sium (as much as 50% of potassium intake) into
the feces. These adaptive processes may be dis-

turbed in SCI patients, with the attendant risk of
life-threatening hyperkalemias. Owing to the
predominant tubulointerstitial nature of renal
disease in SCI patients, adaptive rise in tubular
secretion of potassium (and often H^) may be de-
fective in this population. Moreover, colonic mo-
tility and sphincteric disorders, which frequently
give rise to constipation and fecal impaction in
SCI patients, can interfere with the adaptive co-
lonic potassium secretion. Accordingly, SCI pa-
tients with severe renal insufficiency may be at
greater risk of hyperkalemia than their able-bod-
ied counterparts. This issue should be considered
in the management and monitoring of these pa-
tients, and special attention should be given to
bowel care and impaction control in this setting.

References

1. Hackler RH. A 25-year prospective mortality study in the

spinal cord injured patient: comparison with the long-
term living paraplegic. J Urol 1977; 117:486-488.

2. Borges PM, Hackler RH. The urologic status of the Viet-

nam war paraplegic: a 15-year prospective follow-up. J
Urol 1982; 127:710-711.

3. Price M. Some results of a fifteen year vertical study of

urinary tract function in spinal cord injured patients: a
preliminary report. J Am Paraplegia Soc 1982; 5:31-34.

4. Barton CH, Vaziri ND, Gordon S, Tilles S. Renal pathol-

ogy in end-stage renal disease associated with paraple-
gia. Paraplegia 1984; 22:31-41.

5. Vaziri ND, Cesario T, Mootoo K, Zeien L, Gordon S, Bryne

C. Bacterial infections in patients with chronic renal
failure: occurrence with spinal cord injury. Arch Intern
Med 1982; 142:1273-1276.

6. Vaziri ND. Renal insufficiency in patients with spinal

cord injury. In: Lee, Ostrander, Cochran, Shaw, eds.
The spinal cord injured patient: comprehensive man-
agement. Philadelphia: WB Saunders, 1991:134-157.

7. Vaziri ND, Bruno A, Mirahmadi MK, Golji J, Gordon S,

Byrne C. Features of residual renal function in end-
stage renal failure associated with spinal cord injury.
Int J Artif Organs 1984; 7:319-322.

8. Kaji D, Straus I, Khan T. Serum creatinine in patients

with spinal cord injury. Mt Sinai J Med 1990; 57:160-
164.

9. Mirahmadi MK, Byrne C, Barton CH, Penera N, Gordon

S, Vaziri ND. Prediction of creatinine clearance from
serum creatinine in spinal cord injury patients. Para-
plegia 1983; 21:23-29.
10. Kaw DG, Levy E, Kahn T. Decrease of urine creatinine in
vitro in spinal cord injury patients. Clin Nephrol 1988;
30(4):21fr-219.

Obesity Surgery:

Dietary and Psychosocial Expectations and Reality

J. Gabrielle Rabner, M.S., R.D., Sharron Dalton, PhD, R.D., and Robert J. Greenstein, M.D.

Abstract

Psychosocial and dietary habits have been compared in patients considering (Pre n = 33)
or who had (Post n = 32) surgery for morbid obesity. Failure of conventional diets was
attributed to lack of self-discipline (Pre 92% [22/24] vs. Post 87% [20/23], NS). Consump-
tion of fast foods fell (Pre 70% [23/33] vs. Post 16% [5/32], p < 0.0001). Preoperative
patients had unrealistic social expectations. They exaggerated the prospect of improved
friendship (Pre 67% [22/32] vs. Post 34% [11/21], p < 0.05), erroneously anticipated better
sex (Pre 78% [25/32] vs. Post 50% [15/30], p ^ 0.05), predicted better acceptance at work
(Pre 85% [23/27] vs. Post 50% [15/30], p ^ 0.05), and misanticipated improved relationship
with their partner (Pre 77% [20/26] vs. Post [47% 8/17], p ^ 0.05). Two factors predicted
becoming employed following surgery: age (became employed [n = 5] 28 ± 2 years vs.
remained unemployed [n = 12] 44 ± 4 years, p < 0.05) and percentage of excess weight
lost (became employed 76 ± 11 vs. remained unemployed 51 ± 7, p < 0.05). The free
support group was "useful" (17/17), yet only 5% attended regularly. Patients considering
obesity surgery had specific unrealistic psychosocial expectations. They infrequently
availed themselves of postoperative professional help. We identify the features associated
with gaining emplojmient.

Morbid obesity (1), a major health problem (2),
afflicts seven and one-half million people in the
United States (3). The morbidly obese have obe-
sity-related diseases (2, 4—6), difficulties with
medical insurance (7), and social (8, 9), psycho-
logical (9), and employment-related difficulties
(10). Accumulating evidence suggests that con-
servative dietary therapy does not result in long-
term weight loss for this population (11, 12). As a

From the Bronx VA Medical Center, Bronx, New York (JGR);
New York University, School of Education, Health, and Arts
Professions, New York, NY (SD); and the Department of Sur-
gery, Mount Sinai School of Medicine (CUNY), New York, NY
(RJG). The present address of JGR is Food and Nutrition Ser-
vices, The New York Hospital, 525 East 68th Street, New
York, NY 10021. Address correspondence and reprint requests
to Robert J. Greenstein, M.D., Laboratory of Molecular Sur-
gical Research, VAMC Bronx (151), 130 West Kingsbridge
Road, Bronx, NY 10468.

Presented in part at the 74th annual meeting of the ADA
National Convention, Dallas TX, Oct. 28-Nov. 1, 1991, and
published in a supplement to JADA 1991;91,A25.

result, gastrointestinal surgical therapy for mor-
bid obesity is increasingly being employed (13,
14). A body mass index ^40 is the suggested level
above which surgery may be considered (14).

Vertical banded gastroplasty (VBG) (15) is a
gastric reduction procedure that limits the quan-
tity of food that can be consumed (16, 17). Unlike
some other procedures in current use (18), VBG
has no malabsorptive component. The weight loss
(15, 19-21), preoperative expectation (22), and
physiological (23) and psychological outcome of
VBG (9, 24) have been reported.

The purpose of this study is to compare psy-
chosocial and dietary expectations prior to sur-
gery with the reality that follows VBG. We spe-
cifically addressed fat intake, dietary and weight
changes, subjective impressions, and employment
history. We show that patients often have unre-
alistic expectations and are recalcitrant to long-
term professional follow-up. The likelihood of be-
coming employed following surgery is directly
related to being younger and losing more weight.

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

305

306

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

TABLE 1

Demographic Characteristics

Preoperative Postoperative
(n = 33) (n = 32)

INO.

iNO.

Age (years)

<20

20—29

40—49

Oil— oy

c
O

60-69

Sex

Male

Female

y /

/ /

Marital status

Married

oingle

1 /

6 1

Divorced

Separated

Widowea

Ethnicity*

Black

Hispanic

wnite

Education level

(completed)

None

K-6

Jr high school

High school

College

Graduate school

Employment**

None

Part-time

Full-time

*p « 0.05.
**p ^ 0.01.

Methods

Preoperative (Pre n = 33) and postoperative
(Post n = 32) patients of one surgeon (RJG) were
interviewed either at their initial office visit (Pre)
or by telephone (Post). Postoperative patients
were randomly selected from a list of all 187 post-
surgical patients. All patients who were con-
tacted participated in the study. Data were col-
lected from April 1990 through September 1990.
Patients were interviewed for 30 min (JGR). Fol-
low-up phone calls were made to confirm and
complete information.

Following a pilot study on 10 preoperative
patients, a finalized 175-item questionnaire (ap-
proximately 50 questions) (22) was used. In the
resulting final questionnaire, the subjects ad-

dressed were demographics (8%, n = 14), obesity
history (11%, n = 20), dietary habits (25%, n =
44), weight loss (21%, n = 37), quality of life
(25%, n = 44), and obesity-related diseases (9%,
n = 15). Demographics included age, sex, educa-
tion level, occupation, marital status, and ethnic-
ity.

Dietary habits were obtained through 24-h
dietary recall and cross-checked using reports of
daily and weekly food frequency. This interview
method has been described (25) and shown to
have validity in assessing the average intake of a
group (26, 27). Similar results have been demon-
strated for establishing mean intake for the face-
to-face interview method (used for preoperative
patients) versus the telephone interview method
(used for postoperative patients) (28). Patients
were interviewed in detail about the name and
description of each food item eaten (brand names
when possible), preparation method, and portion
size, using household measures (teaspoons, table-
spoons, cup sizes). These measures have been
shown to enhance the ability of the interviewee
to estimate food portion sizes accurately (29).
Weight history over the year preceding the inter-
view and expectations of weight loss maintenance
were addressed. "Support systems" items identi-
fied the professional or nonprofessional person
sought when dietary or related problems arose.
"Quality of life" items addressed health, personal
appearance, employment, social relationships,
and self-esteem.

Certain aspects of this study have been re-
ported elsewhere (22). Included were obesity his-
tory, health-related aspects of obesity, weight
loss, and its maintenance.

Statistical Analyses. Data from the com-
pleted questionnaires were entered into Excel
(Microsoft, Redmond, WA) and Statview 512 +
(Abacus Concepts) using Mac Plus Computer (Ap-
ple Computer, Cupertino, CA). Percentage of
ideal body weight was calculated using the 1983
Metropolitan Life Tables (medium frame). Di-
etary analysis was performed using Nutritionist
III (Analytic Software, Salem, OR). The database
of this program contains 2,025 foods and 58 nu-
trients. The 24-h dietary recall information was
analyzed for grams and percentage of carbohy-
drate, protein, fat, and saturated fat. These were
compared to the Dietary Guidelines for Healthy
American Adults of the American Heart Associa-
tion (30). Statistical analysis was by chi-squared,
unpaired ^-tests or parametric ANOVA as appro-
priate. Data are presented as mean ± SEM. Sig-
nificance was accepted at p < 0.05.

Vol. 60 No. 4

OBESITY SURGERY— RABNER ET AL.

307

Results

Sample. The preoperative subjects were in-
terviewed at their initial office visit. The postop-
erative group were interviewed from 4 to 84
months (24 ± 4 mean ± SEM) following surgery.
Similarities in demographic characteristics were
found in male/female ratios, age, educational lev-
els, marital status, and presurgery weight of both
groups. Weight (Pre 287 ± 9 lb vs. Post 291 ± 13
lb, NS), body mass index (Pre 49.6 ±1.6 vs. Post
48 ± 1.7, NS), and percent of ideal body weight
(Pre 218 ± 6 vs. Post 211 ± 6 NS) were similar.
Males were heavier than females (&225% of IBW:
males 67% [4/6] vs. females 34% [20/59], p «
0.01).

A large percentage of the population was His-
panic (48%) and female (89% ). There were signif-
icantly more females than males in both groups
(Table 1). The mean age was 36 years (Pre 33 ±
1.6 years vs. Post 38.6 ± 2.6 year, NS), with 8%
(5/65) of patients below the age of 20, and 11%
(7/65) of patients above the age of 50. Statistical
differences between groups were observed in eth-
nicity and employment (Table 1).

Nutrition. Weight regain following conven-
tional diets was ascribed by patients to lack of
self-discipline (Pre 92% [22/24] vs. Post 87% [20/
23], NS). The absence of support systems was ad-
ditionally implicated as a reason for failure (Pre
70% [16/23] vs. Post 54% [14/26], NS). A minority
of patients had consulted a dietitian before re-
questing surgery (Pre 16% [5/32] vs. Post 6%
[2/32], NS).

Preoperative diets were higher than recom-
mended guidelines in simple sugars, processed
meats, and fast food products (30). Diets con-
tained ^20% of recommended complex carbohy-
drates and fiber (average 5 g in both groups) (30).
Postoperative patients generally consumed the
same types of food but in greatly reduced quanti-
ties (22). Foods high in fat, such as cake, cookies,
and ice cream, remained favorite items for both
groups.

Patients' caloric intake fell significantly fol-
lowing surgery (calories: Pre 5,283 ± 356 vs. Post
1,622 ± 95, p ^ 0.001) (22). No difference was
found in the proportion of fat intake (% fat: Pre
40% vs. Post 37%, NS). The number of subjects
eating fast foods fell precipitously following sur-
gery (Pre 70% [23/33] vs. Post 16% [5/32], p <
0.0001).

The dietary differences between patients who
lost <49.9% or ^50% of their excess body weight
following surgery are presented in Table 2. Dif-
ferences in protein between the groups may result

TABLE 2

Analysis of 24-hour Dietary Recall of Postoperatiue Patients
by % of Excess Weight Loss

Excess

wt loss

s49.9%

«50%

(pt no. 21)

(pt no. 9)

Protein (g)

55 ± 6

80 ± 12

«0.05

Carbo-

hydrate (g)

187 ± 14

226 ± 30

Fat (g)

66 ± 7

81 ± 8

Total

calories

1,500 ± 106

1,800 ± 183

from the decrease in total calories. The postoper-
ative group preferred diet soft drinks. Other dif-
ferences in nutritional choices between the two
groups were negligible (data not presented).

Weight loss of the postoperative group was
104 ± 14 lb (22). The preponderance of weight
loss occurred in the first year following surgery,
which is in agreement with previous studies (19,
31, 32) (actively losing: 0-12 months [n = 14]
100% vs. 13-84 months [n = 18] 42%, p ^ 0.01).
Patients thought they would maintain weight
loss following surgery (Pre 97% [32/33] vs. Post
94% [30/32], NS). In reality, 44% of postoperative
patients were actively losing weight, and 50%
had stabilized their weight. Postoperative pa-
tients said that they would not be able to control
their weight if surgery were reversed (Post 86%
[26/31]).

Both groups "strongly agreed" that their goal
was to lose 50% of excess weight in one year fol-
lowing surgery (Pre 97% [32/33] vs. Post 97% [29/
30], NS). Sixty-six percent [21/32] of postopera-
tive patients lost and maintained &50% of their
excess body weight over a 4— 84— month period
(22).

Employment. More individuals were em-
ployed in the postoperative group (Pre 30% [10/
33] vs. Post 56% [18/32], p < .05) (Table 1). Two
factors identified those who became employed fol-

TABLE 3

Postoperative Employment Changes

Became Remained
employed unemployed

No. of patients

5 (29%)

12 (71%)

Mean age

28 ± 2.4*

44 ± 4.4

Male/female

1/4

3/9

Time postop, months

10 ± 2

13 ± 4.4

% excess wt loss

76 ± 11*

51 ± 7

Mean ± SEM. *p « 0.05.

17 patients unemployed prior to antiobesity surgery.

308

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

100
80

^ 60 -

o
a:

Q. 40 -

20 -

I 1 Preoperalive
188883 Postopetalive
p=NS

Social Events

Social
Acceptance

Q- 40

20 -

30/32 30/33

Live Longer
Lite

Weight Loss
Maintenance

Se«
ConliOence

Fig. 1. Elements of quality of life on which expectations of preoperative patients were relatively realistic as compared to
expectations of postoperative patients.

lowing surgery (Table 3). They were younger (be-
came employed [n = 5] 28 ± 2 years vs. remained
unemployed [n = 12] 44 ± 4 years, p < 0.05) and
had lost more weight (% excess weight lost: be-
came employed 76 ± 11 vs. remained unemployed
51 ± 7, p < 0.05).

Quality of Life. Concordance between preop-
erative and postoperative group expectations was
evident on many aspects of quality of life (Fig. 1).
By contrast, preoperative patients had unrealis-
tically high expectations, as compared to the ex-
pectations of postoperative patients, for accep-
tance at work, ability to make and keep friends,
better sex, and better relationships (Fig. 2).

Support Systems. A professionally guided
monthly support group session is available free to
all patients. Twenty-eight percent of preopera-
tive patients and 53% of postoperative patients
had attended at least once. Regular attendance
among postoperative patients is 5%. Of the 17 pa-
tients who had attended, all said it had been help-
ful. This study did not address the reasons for the
limited attendance.

Following surgery, there was an increase in
family and friends who approved of the operation.
Family (Pre 79% [27/33]) and friends (Pre 88%
[29/33]) had been told about the proposed surgery.
Only 57% (18/33) of family and 52% (17/33) of
friends approved. Six representatives of disap-
proving family and friends, when recontacted fol-
lowing surgery (87% [6/7]), had become support-
ive.

Discussion

This study provides insights for the bariatric
surgeon and other health professionals to better
prepare the patient for the reality of life following

surgery. We document an improved quality of life
and better employment prospects following sur-
gery for morbid obesity. Postoperatively, these
patients are not availing themselves of accessible
professional dietary counseling. All patients who
have attended the supervised monthly support
group attest to its usefulness. Yet only ^15% do
so. We identify the critical period, one year fol-
lowing surgery, when education may be optimally
provided. We emphasize reducing fat intake in
our postoperative counseling.

In many areas, patients have realistic expec-
tations (Fig. 1); however, we find they are unre-
alistic about sex, friends, and acceptance at work
(Fig. 2). Our subjects were recruited in a random
manner. Follow-up studies to validate these ob-
servations may be conducted on either the same
individual prior to and following surgery or pair-
matched individuals.

Unsatisfactory weight loss following gastric
restrictive surgery has been ascribed to excess
"sweet" consumption (19). It is possible that fat is
of equal or greater importance. Body fat is di-
rectly related to fat intake (33), and morbidly
obese people may be especially efficient at lipid
storage (34). Our patients' consumption of fat is
similar to that of the general American popula-
tion (35). Fat comprises —69% of calories from
cake, —70% of calories from candy, and —48% of
calories from ice cream (36). Patients may be un-
aware of the need to decrease their fat intake (30),
unaware of the hidden grams of fat they are con-
suming, or unwilling to change their dietary hab-
its. The continued high consumption of fat may be
due, in part, to the palatability of fat foods and
the satiety fat induces (37).

Body weight may be decreased by lowering
dietary fat without restricting quantity of food

Vol. 60 No. 4

OBESITY SURGERY— RABNER ET AL.

309

100

:60

40-

I I Preoperative
1 I Postoperative

Relationship
with partner
p=<005

Better sex
p.<0,05

Acceptance
at work
p.<005

Ability to make
and keep friends
p=<0 01

Fig. 2. Elements of quality of life on which expectations of
preoperative patients diverged widely from expectations of
postoperative patients.

consumed (38). We show that postoperative pa-
tients continue to consume a diet high in fat (Ta-
ble 2). Our postoperative counsehng now empha-
sizes decreasing fat intake to ^40 g/day. At the
initial preoperative office visit, patients are given
an informational pamphlet to assist in this aim.
After surgery, they receive a reference book that
itemizes the fat content of most foods (39).

There is an association between ethnic origin
and obesity (40). Obesity-related diseases are
prevalent in medical clinics used by Hispanics (7).
In this study, a large percentage of the patient
population is Hispanic. We ascribe this to the eth-
nic mixture in the New York metropolitan area,
as well as to patient-patient referral patterns. In
the preoperative group they constitute the major-
ity (64%, 21/33), whereas in the postoperative co-
hort their representation falls to 31% (10/31) (Ta-
ble 1). This may have resulted in biased results.
However, an ethnically stratified analysis of di-
etary habits did not show any differences (data
not presented). This may mean that the popula-
tion, self-selected by weight, do indeed have sim-
ilar dietary habits. Alternatively, in order to
identify ethnic-dependent differences, a larger
number of subjects will need to be studied.

In the United States the prevalence of obesity
in men and women is about equal (9). We confirm
that a disproportionate percentage (41) of those
attempting to control their weight are females
(and see the New York Times, Nov. 24, 1991). We
show that when men do seek help, they are heavi-
er than women. Social stigma may, in part, en-
courage more females to seek surgical weight con-
trol assistance (42). Our patients view lack of self-
discipline as the most important factor for not
maintaining previous weight loss. Although mul-
tiple attempts at dieting had been made, for some

of the patients perioperative dietary counseling
by our registered dietitian may be their first pro-
fessional contact. However, this deficiency can-
not be ascribed solely to the lack of opportunity
to associate with professionals, as only 5% of
patients attend the free, professionally run,
monthly support group. We ascribe the lack of
support of family and friends prior to surgery pre-
dominantly to fear of death at surgery.

Postoperative patients stated that they are
consuming 25% of the calories they ate prior to
surgery (22). However, the ratio between carbo-
hydrates, protein, and fat did not change. Nonsig-
nificant changes in consumption of these nutri-
ents before and after surgery have been observed
in other studies (43). Patients' dietary mix is low
in dietary fiber and high in fat (22).

Remarkably, postoperative patients have
vastly reduced the total amount of fast foods con-
sumed. This may be a positive change in dietary
habits. More probably, it may be ascribed to the
enforced behavior modification, vomit avoidance,
induced by the operation (22).

Patients have unrealistic goals for long-term
weight loss (22). We confirm that weight loss is
maximal in the first year following surgery (19,
32). Success in maintaining weight loss may, in
part, depend on continuing dietary and/or other
support systems (44, 45). Nevertheless, our re-
sults concur (14, 15, 19, 46) that surgery is the
optimal therapy for this population at present.

Our patient populations confirm a very high
unemployment rate in the population of the mor-
bidly obese (10). This may be ascribed to reluc-
tance to hire the morbidly obese (10) and to poor
body image (47). Societal prejudice may also play
a role (42). Improvement at work follows vertical
banded gastroplasty (41, 48). We show that pa-
tients who became employed were younger and
had lost a higher percentage of their excess body
weight than those who remained unemployed.
The implications to society of an efiective treat-
ment for the young obese population who are on
welfare is of obvious importance.

References

1. Build study. Chicago: Society of Actuaries and Associa-

tion of Life Insurance Medical Directors of America,
1983.

2. Task Force of the American Society for Clinical Nutrition.

Guidelines for surgery for morbid obesity. Am J Clin
Nutr 1985; 42:904-905.

3. Kuczmarski RJ. Prevalence of overweight and weight

gain in the United States. Am J Clin Nutr 1992; 55:
495S-502S.

4. Stunkard AJ, Stinnett JL, SmoUer JW. Psychological and

310

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

social aspects of the surgical treatment of obesity. Am J
Psychiatry 1986; 143:417-429.

5. Bray GA. Complications of obesity. Ann Intern Med 1985;

103:1052-1062.

6. Krai JG. Morbid obesity and related health risks. Ann

Intern Med 1985; 103:1043-1046.

7. Council on Scientific Affairs. Hispanic health in the

United States. JAMA 1991; 265:248-252.

8. Van Italie TB, Krai JG. The dilemma of morbid obesity.

JAMA 1981; 246:999-1003.

9. Wadden TA, Stunkard AJ. Social and psychological con-

sequences of obesity. Ann Intern Med 1985; 103:1062-
1067.

10. Howard L, Jenks JS. Medical aspects of morbid obesity.

In: Linner JH, ed. Surgery for morbid obesity. New
York: Springer- Verlag, 1984: 1-16.

11. Technology Assessment Conference Panel. Methods for

voluntary weight loss and control. Technology Assess-
ment Conference statement. Ann Intern Med 1992; 116:
942-949.

12. Leon GR. Personality and morbid obesity. Surg Clin

North Am 1979; 59:1007-1015.

13. Owen ERTC, Kark AE, Cooper MK. Obesity surgery in

the United Kingdom: survey of 970 general surgeons.
Ann R Coll Surg Engl 1991; 73:36-38.

14. Gastrointestinal surgery for severe obesity: National In-

stitutes of Health Consensus Development Conference
Statement. Am J Clin Nutr 1992; 55:615S-619S.

15. Mason EE. Use of vertical banded gastroplasty. Surg Clin

North Am 1987; 67:521-537.

16. McFarland RJ, Ang L, Parker W, Pilkington TRE, Gazet

JC. The dynamics of weight loss after gastric partition
for gross obesity. Int J Obesity 1988; 13:81-88.

17. Villar HV, Wangensteen SL, Burks TF, Patton DD. Mech-

anisms of satiety and gastric emptying after gastric
partitioning and bypass. Surgery 1981; 90:229-236.

18. Holian DK. Biliopancreatic bypass. In: Deitel M, ed. Sur-

gery for the morbidly obese patient. Philadelphia: Lea
& Febiger, 1989: 105-111.

19. Sugarman HJ, Starkey JV, Birkenhauer R. A randomized

prospective trial of gastric bypass versus vertical
banded gastroplasty for morbid obesity and their effects
on sweets versus non-sweets eaters. Ann Surg 1987;
205:613-622.

20. Lechner GW, Elliott DW. Comparison of weight loss after

gastric exclusion and partitioning. Arch Surg 1983;
118:685-692.

21. Brolin RE. Critical analysis of results: weight loss and

quality of data. Am J Clin Nutr 1992; 55:577S-581S.

22. Rabner JG, Greenstein RJ. Obesity surgery: expectation

and reality. Int J Obesity 1991; 150:841-845.

23. Jones BA. Vertical banded gastroplasty. In: Deitel M, ed.

Surgery for the morbidly obese patient. Philadelphia:
Lea & Febiger, 1989: 145-165.

24. Halmi KA, Stunkard AJ, Mason EE. Emotional responses

to weight reduction by three methods: gastric bypass,
jejunoileal bypass and diet. Am J Clin Nutr 1980; 33:
446-451.

25. Burke BS. The dietary history as a tool in research. JADA

1947; 23:1041-1046.

26. Gersovitz M, Madden JP, Smiciklas- Wright H. Validity of

the 24-hr dietary recall and seven-day record for group
comparisons. JADA 1978; 73:48-55.

27. Block G. A review of validations of dietary assessment

methods. Am J Epidemiol 1982; 115:492-505.

28. Posner BM, Borman CL, Morgan JL, Borden WS, Ohis JC.

The validity of a telephone-administered 24-hour di-
etary recall methodology. Am J Clin Nutr 1982; 36:
546-553.

29. BoUand JE, Yuhas JH, Bolland TW. Estimation of food

portion sizes: effectiveness of training. JADA 1988; 88:
817-821.

30. Nutrition Committee of the American Heart Association.

Dietary guidelines for healthy American adults. Dallas:
American Heart Association, 1988.

31. Deitel M, ed. Surgery for the morbidly obese patient. Phil-

adelphia: Lea & Febiger, 1989.

32. GrifTin WO, Printen KJ. Surgical management of morbid

obesity. New York: Marcel Dekker, 1987.

33. Dreon DM, Frey-Hewitt B, Ellsworth N, Williams PT,

Terry RB, Wood PD. Dietary fat: carbohydrate ratio
and obesity in middle-aged men. Am J Clin Nutr 1988;
47:995-1000.

34. Roncari DAK, Angel A. The fat cell. In: Deitel M, ed.

Surgery for the morbidly obese patient. Philadelphia:
Lea & Febiger, 1989: 3-18.

35. The Surgeon General report on nutrition and health.

Washington, DC: Government Printing Office, DHHS
(PHS) Publication no. 88-50210, 1988.

36. Pennington JAT, Church HN. Food values of portions

commonly used, 15th ed. New York: Harper & Row,
1989.

37. Expert Panel on Food Safety and Nutrition. Fats in the

diet: why and where? Food Tech 1986; October: 115-
120.

38. Kendall A, Levitsky DA, Strupp BJ, Lissner L. Weight

loss on a low-fat diet: consequence of the imprecision of
the control of food intake in humans. Am J Clin Nutr
1991; 53:1124-1129.

39. Natow AB, Heslin JA. The fat counter. New York: Simon

& Schuster, 1989.

40. Pawson IG, Mortorell R, Mendoza FE. Prevalence of obe-

sity in U.S. Hispanic population. Am J Clin Nutr 1991;
Suppl 53:1522s-1528s.

41. Brolin RE. Results of obesity surgery. Gastroenterol Clin

North Am 1987; 16(2):317-338.

42. Czajka-Narins DM, Parham ES. Fear of fat: attitudes to-

ward obesity. Nutr Today 1990; January/February:26-
32.

43. Naslund I, Jarnmark I, Anderson H. Dietary intake be-

fore and after gastric bypass and gastroplasty for mor-
bid obesity in women. Int J Obesity 1988; 12:503-513.

44. Hall JC, Watts JM, O'Brien PE, et al. Gastric surgery for

morbid obesity. Ann Surg 1990; 211:419-427.

45. Bufalino J, Kolisetty L, McCaskey GW, Stratton KL. Sur-

gery for morbid obesity: the patients' experiences. Appl
Nurs Res 1989; 2:16-22.

46. Yale CE. Gastric surgery for morbid obesity. Arch Surg

1989; 124:941-947.

47. Stunkard AJ. The pain of obesity. Menlo Park, CA: Bull

Publishing Co, 1976.

48. Powers PS, Rosemurgy AS. Psychological sequelae of sur-

gical procedures for obesity. In: Deitel M, ed. Surgery
for the morbidly obese patient. Philadelphia: Lea & Fe-
biger, 1989: 351-358.

Submitted for publication July 1992.
Final revision received February 1993.

Phase II Trial of Etoposide, Carboplatin,
and Ifosfamide as Salvage Therapy in
Advanced Ovarian Carcinoma

Ann Marie Beddoe, M.D., Peter R. Dotting, M.D., and Carmel J. Cohen, M.D.

Abstract

A phase II study combining etoposide with carboplatin and ifosfamide as salvage therapy
in advanced ovarian cancer was undertaken. Objective responses were achieved in 37.5%
of 16 evaluable patients with a mean progression-free interval of 8.6 months. Stable
disease was present in 25% of patients; in 37.5% of patients the disease progressed on
salvage. Based on original response to front-line therapy, patients were classified as being
platinum-sensitive (group I) or platinum-refractory (group II). Clinical response to salvage
therapy was seen in 44.5% of group I patients, but in only 28.6% of group II patients. This
difference was not statistically significant. When a more precise definition of platinum
sensitivity was applied, clinical responses were seen in 54.5% of group I patients, but no
responses were noted among group II patients ip < 0.05). Platinum sensitivity appeared to
be an important factor in achieving a response with this regimen. This combination was
well tolerated, myelotoxicity being the dose-limiting toxicity encountered. No life-threat-
ening, nonhematologic toxicities were seen. One death occurred secondary to nadir sepsis.
The combination of etoposide, carboplatin, and ifosfamide is an active salvage regimen in
patients with advanced ovarian carcinoma; however, severe myelotoxicities and inability
to produce long-term responses underscore the need for continued trials to find a more
durable salvage regimen.

The use of platinum combination therapy in pa-
tients with advanced stage III and IV ovarian
cancer produces clinical response rates of greater
than 75% (1-3). Surgical complete responses,
however, fall below 40% (1, 4, 5), and the outlook
for patients with >2 cm residual disease is re-
stricted (6). The relapse rate among patients who
achieve a complete response by pathologic exam-
ination is estimated at 45%-65%, most of these
relapses occurring within two years of second-
look laparotomy (7).

Several regimens have been proposed as sal-
vage therapy for those patients whose disease
progresses during front-line treatment or in

From the Division of Gynecologic Oncology, Department of
Obstetrics, Gynecology, and Reproductive Science, The Mount
Sinai Medical Center, One Gustave L. Levy Place, Box 1173,
New York, NY 10029. Address reprint requests to Ann Marie
Beddoe, M.D., at Mount Sinai.

whom carcinoma recurs following successful
treatment for advanced ovarian cancer. These
regimens have resulted in clinical responses of
10%-50%, with progression-free intervals rang-
ing from 6 to 12 months (8—11). The optimal sal-
vage regimen, however, has not been defined, be-
cause to date no regimen has produced durable
responses. Current investigations are therefore
focusing on second-generation analogs of active
first-line agents. Among the analogs that have
been tested, carboplatin (JM8), a cisplatin analog,
and ifosfamide, an analog of cyclophosphamide,
have emerged as agents demonstrating therapeu-
tic activity in patients with advanced ovarian car-
cinoma (12, 13).

The podophyllotoxin derivative etoposide has
also been evaluated as a single agent in the treat-
ment of advanced ovarian cancer, producing clin-
ical response rates of 22% (14), and it appears to
be synergistic with cisplatin, as demonstrated
with L1210 leukemia cells (15). Carboplatin, a

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

311

312

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

TABLE 1

Characteristics of 17 Patients

No. of
pts

Age

Median

56.8 yr

Range

45-50 yr

Prior therapy

CAP

C-IP cisplatin

Cisplatin

Mean

800 mg

Range

465-1650 mg

Prior salvage

therapy, range

1-3 courses

AP, adriamycin, cisplatin; CP, Cytoxan, cisplatin; CAP, Cy-
toxan, adriamycin, cisplatin; C-IP, cisplatin, intravenous
high-dose Cytoxan, intraperitoneal cisplatin.

cisplatin analog, has been found to be as effective
as cisplatin in the primary treatment of ovarian
cancer and has been proposed as the first choice in
retreatment for patients who have had a previous
response to cisplatin (16). As primary treatment
for advanced stages III and IV ovarian cancer,
ifosfamide produces responses of 33% (17); how-
ever, its importance in salvage regimens may lie
in its ability to produce responses in patients with
platinum-refractory ovarian cancer (18).

Combining three agents with known individ-
ual activity in advanced ovarian carcinoma, we
undertook a phase II study utilizing etoposide
with carboplatin and ifosfamide as salvage ther-
apy for patients with advanced ovarian carci-
noma that has recurred following successful
treatment or has progressed during treatment
with platinum-containing agents.

Materials

Seventeen patients with stages III and IV
ovarian cancer who had received primary treat-
ment with a platinum-containing regimen follow-
ing initial surgery were entered on a phase II
trial utilizing etoposide, carboplatin, and ifosfa-
mide as a salvage regimen. The protocol was ap-
proved by the Institutional Review Board of the
Mount Sinai Medical Center, and informed con-
sent was obtained from all patients prior to ad-
ministration of chemotherapy.

Eligibility criteria included histologically
proven advanced epithelial cancer and pretreat-
ment with a platinum-based regimen. All pa-
tients had an expected survival of greater than 12
weeks and Karnofsky performance status of at
least 40. Measurable disease was present in all
patients as determined by measurements on com-
puted-tomography scans. Baseline hematologic
data for entry into this protocol included white
blood cell count more than 3500/mm^ and platelet

TABLE 2

Status Following First-Line Cisplatin Therapy

Pt. Second Time to Site of Response

no. Stage look recur recurrence to salvage

1
2
3
4
5
6
7
8
9
10

1
2
3
4
5
6
7

Group I platinum sensitive

Negative 24 mo

Negative 17 mo

Negative 28 mo

Refused 38 mo

Micro + 10 mo
Micro + 7 mo

Micro -I- 11 mo

Micro -I- 20 mo

Group II platinum refractory
None 6 mo

None 6 mo

None Progression

None Progression
None 3 mo

None 4 mo

Abdomen

Pelvic l.n.

Carcino

Liver

Pelvis

Carcino

Pelvis

Carcino

Porta hep

Abdomen
Pelvic l.n.
Pelvis

Liver, ascites
Pelvis, abd
Pelvis

Liver, carcino

CR X 4 mo
CR X 5 mo
PR X 5 mo
Stable X 5 mo*
Dead — Tx comp.
CR X 4 mo
Progression*
Progression*
Stable X 3 mo*
Stable X 5 mo*

CR X 5 mot
CR X 4 mo

Progression*!
Progression
Progressiont
Progression
Stable X 4 mo*t

Micro, microscopic disease; abd, abdomen; CR, complete response; PR, partial response; l.n., lymph nodes; carcino, carcinomatosis;
porta hep, porta hepatus.

* Treatment with at least one salvage regimen before entering on protocol,
t Secondary debulking attempted.

Vol. 60 No. 4

SALVAGE THERAPY IN OVARY CA— BEDDOE ET AL.

313

counts greater than 125,000/mm^. Normal he-
patic function and serum creatinine less than 1.5
mg/dL were also required.

The dosages for these three agents were as
follows: ifosfamide, 1.2 g/m^ days 1-3; etoposide,
100 mg/m^ days 1-3; carboplatin, 100 mg/m^ days
1-3.

The uroepithelial protector sodium-2-mer-
captoethane sulfonate (MESNA) was adminis-
tered each day that ifosfamide was given, with
the aim of preventing hemorrhagic cystitis (19).
MESNA was administered intravenously in three
equally divided doses prior to administration of
ifosfamide and repeated at four hours and eight
hours following ifosfamide infusion. All patients
were admitted for the duration of the chemother-
apy. Toxicities were graded according to the Gy-
necologic Oncology Group criteria for toxicities.
Dose reductions were made for grade 3-4 toxici-
ties as follows: 10% reduction if nadir counts of
platelets and white blood cells were 25,000-
50,000/mm^ and 500-1, 000/mm^, respectively,
and 25% reduction if platelet count was less than
25,000/mm^ and white blood count less than 50/
mm^.

Treatment was delayed more than four
weeks until platelet counts increased to at least
100,000/mm^ and white blood cell counts to 3,000/
mm^ or greater.

Responses were defined as complete (CR) if
there was disappearance of all physiologic and ra-
diologic evidence of disease, with sustained ab-
sence of CA-125 levels where applicable, or as
partial (PR) if there was a decrease greater than
50% in the product of the two largest perpendic-
ular dimensions of existing lesions in the absence
of new lesions. The disease was defined as stable

TABLE :}

Survival of Patients Treated with Salvage Therapy

Mean

No. of

Response

survival

No. pts

pts

to salvage

(range)

alive (%)

1 PR, 5 CR

14.3 mo

4 (66.7)

(9-23+ mo)

Stable dz

15.5 + mo

2 (50.0)

(7-23+ mo)

Progression

7.7 mo

0(0.00)

(4-12 mo)

PR, partial response; CR, complete response; stable dz, stable
disease.

One patient died of treatment-related complications.

if there was no evidence of new tumor growth or
less than 25% increase in the product of the two
largest dimensions of an existing tumor mass or
the onset of new effusions and lesions.

All patients were followed up with weekly
laboratory data, monthly examinations, and ra-
diologic studies when indicated.

Results

All 17 patients entered on this study were
evaluable for toxicity and response (Table 1). All
patients were previously treated for either stage
III or stage IV ovarian cancer.

Two groups of patients were identified in the
study population, based on response to first-line
cisplatin therapy (Table 2). Group I consisted of
patients who demonstrated a clinical response to
first-line cisplatin treatment (platinum sensi-
tive). Patients were considered to demonstrate
clinical platinum sensitivity if they were clini-
cally free of disease at the end of therapy and

TABLE 4

Comparison of Response to Salvage Therapy of Platinum-Sensitive and Platinum-Refractory Patients Using Original and

Revised Definitions of Sensitivity

Clinical

response Stable
(CR + PR) dz

progression

Group I: sensitive (9 pts)*

Original Definitioni:

4 pts (44.5%) 3 pts (33.3%)

2 pts (22.2%)

Group II: refractory

2 pts (28.6%) 1 pt (14.3%)

4 pts (57.1%)

Group I: sensitive (11 pts)*

Revised Definitioni

6 pts (54.5%) 3 pts (27.3%)

2 pts (18.2%)

Group II: refractory

0 pts (0.00%) 1 pt (20.0%)

4 pts (80.0%)

.05

CR, complete response; PR, partial response; Dz, disease.

* 1 patient died of treatment-related complications.

t Statistical testing using x"^ test of independence df = 2, p-value <.05.

t Original definition of platinum sensitivity: patients who demonstrated clinical response to first-line cisplatin by being clinically
free of disease at end of therapy and were eligible for second-look laparotomy. Revised definition: patients who, after treatment
with first-line platinum, are clinically and chemically free of disease and remain so for up to 6 months.

314

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

were eligible for second-look laparotomy. Ten
platinum-sensitive patients (59%) were identi-
fied, nine of whom underwent a second-look lap-
arotomy.

Group II consisted of patients who were plat-
inum refractory. All demonstrated either progres-
sion of disease during therapy or measurable re-
currence within six months of completion of a
platinum-containing regimen. Seven patients
(41%) were determined to be platinum refractory.

Response to Therapy. Table 3 shows overall
response to therapy among 16 of the patients who
were eligible for evaluation of response. Table 4
compares the responses between group I (plati-
num sensitive) and group II (platinum refractory)
patients.

Hematologic Toxicity. A total of 12 patients
(70.6%) required at least one dose reduction for
hematologic toxicity. Fifty percent (6 patients)
experienced severe thrombocytopenia, with a
mean nadir platelet count of 27.7 K (range 3
K-49 K). In 45% (5 patients), therapy was com-
plicated by severe neutropenia. Three of these pa-
tients were treated for granulocytopenic fevers.
In one, sepsis was documented, and despite ag-
gressive medical management in an intensive
care unit, the patient succumbed to infection.
Blood transfusions were administered to 8 pa-
tients; hemoglobin counts below 6.5 g/fxL were
seen in 2 patients.

Nonhematologic Toxicity. All patients expe-
rienced grade 3 alopecia; however, gastrointesti-
nal, genitourinary, and neurologic toxicities were
all mild. In no patient was therapy discontinued
because of any nonhematologic toxicity.

Discussion

This clinical trial was designed to investigate
the efficacy of the combination of etoposide, car-
boplatin, and ifosfamide for use as salvage ther-
apy in advanced ovarian carcinoma. Alone, each
agent has produced clinical responses of
22%— 50% as first-line therapy in patients with
advanced ovarian carcinoma (14, 16, 17). The
mechanism of action and toxicities of these agents
vary, and although various combinations utiliz-
ing one or more of these drugs have been re-
ported, this is the first report combining these
three agents in the treatment of recurrent or per-
sistent ovarian carcinoma.

Etoposide exerts its cytotoxic effect by inhib-
iting DNA synthesis through the formation of a
complex with DNA and topoisomerase II, the
DNA-unwinding enzyme. As such, it causes the
arrest of cells in the late S or early G2 phase (20,

21). Its only significant side effect is myelotoxic-
ity; thus it can be combined with other cytotoxic
agents and be administered safely to heavily pre-
treated patients. When etoposide has been com-
bined with cisplatin as salvage therapy in ovari-
an cancer, clinical response rates of 53% have
been demonstrated (11).

Carboplatin, like its parent cisplatin, binds
covalently to DNA, thus disrupting DNA func-
tion. In the treatment of ovarian cancer, both
agents are equally effective; however, carboplatin
has a spectrum of toxicities that differs from that
of cisplatin, resulting in much less nephrotoxic-
ity, neurotoxicity, and emesis (22, 23). It is estab-
lished that as salvage therapy, cisplatin will pro-
duce responses in alkylating-agent-resistant
ovary cancer, in patients who have had intensive
prior therapy, and in patients demonstrating pre-
vious sensitivity to cisplatin (24-26); however, se-
vere toxicities on the renal and neurological sys-
tems have limited its use. Like cisplatin,
carboplatin is also effective in the treatment of
platinum-sensitive ovary cancer. Although clini-
cal studies indicate cross-resistance between
these two drugs, it may be incomplete; responses
to carboplatin in cisplatin-refractory ovary cancer
have been demonstrated in approximately 5% of
patients (27). Ifosfamide appears to be a reason-
able alternative in the management of these pa-
tients.

In this study, the 28% clinical response rate
seen among the cisplatin-refractory individuals
at first glance appears to support incomplete
cross-resistance between carboplatin and cisplat-
in. This response rate is also higher than that
reported by Ozols et al and Colombo et al, who
saw no responses to high-dose carboplatin among
patients who failed to respond to cisplatin regi-
mens (28, 29). In an attempt to explain this dis-
crepancy, we reviewed the responses among the
cisplatin-refractory patients in this study and
found that the only responses in this group were
in patients whose measurable disease occurred
six months following first-line therapy. It appears
from this observation that patients with recur-
rences at six months following first-line cisplatin
behave more like cisplatin-sensitive individuals.
We therefore propose a definition of platinum-
sensitive patients as those who, following treat-
ment with first-line platinum, are clinically and
chemically free of disease and remain so for up to
six months following first-line therapy. The im-
portance of a precise definition of platinum re-
sponse among ovarian cancer patients has re-
cently been addressed (30).

Using our new definition, we compared the

Vol. 60 No. 4

SALVAGE THERAPY IN OVARY CA— BEDDOE ET AL.

315

response to salvage therapy between platinum-
sensitive and platinum-refractory patients and
found that the difference between these groups
was statistically significant (Table 4). The ad-
justed data also appear more compatible with
Ozols' findings. The benefit of ifosfamide in pro-
ducing responses among patients whose disease
progresses on cisplatin is not seen in this small
group of patients.

Using this combination of etoposide, carbo-
platin, and ifosfamide in patients previously
treated with at least one platinum-containing
regimen, we have demonstrated that the dose-
limiting toxicity is clearly hematologic; this has
been similarly demonstrated with the use of car-
boplatin in previously treated patients (29). Seri-
ous hematologic toxicity resulting in dose reduc-
tion occurred in over 70% of patients, and
infection secondary to severe myelosuppression
was present in over 5% of patients. Neurotoxicity,
in particular somnolence, confusion, and aggra-
vation of preexisting psychiatric conditions, has
been reported in approximately 5% of patients re-
ceiving ifosfamide (31). None of these effects were
observed in this population, even among those pa-
tients with psychiatric disease that predated en-
try into this protocol.

The present study indicates that when used
in combination as salvage treatment for advanced
ovarian cancer, etoposide, carboplatin, and ifosfa-
mide produce clinical response rates of over 35%,
with a mean progression-free interval of 8.6
months. The combination appears to be beneficial
in patients who have demonstrated previous plat-
inum sensitivity. These results demonstrate that
this combination has activity as a salvage regi-
men in advanced ovarian cancer, and in a well-
selected population can equal the clinical re-
sponse rates of 53% seen with the use of etoposide
and cisplatin, as previously reported from this in-
stitution. The associated morbidity and mortality
with this regimen are, however, higher than with
etoposide and cisplatin alone. Continued trials,
using other triple combinations in an attempt to
achieve a more durable response while decreasing
morbidity and mortality, will therefore continue.
We stress the importance of identifying those pa-
tients who demonstrate previous platinum sensi-
tivity, because these patients appear to benefit
most from retreatment with platinum analogs.

References

1. Greco FA, Julian CG, Richardson RL, Burnett L, Hande

KR, Oldham RK. Advanced ovarian cancer: brief inten-
sive combination chemotherapy and second-look opera-
tion. Obstet Gynecol 1981; 58:199-205.

2. Cohen CJ, Goldberg JD, Holland JF, et al. Improved ther-

apy with cisplatin regimens for patients with ovarian
carcinoma (FIGO stages III and IV) as measured by
surgical end-staging (second-look operation). Am J Ob-
stet Gynecol 1983; 145:955-967.

3. Omura G, Blessing JA, Ehrlich CE, et al. A randomized

trial of cyclophosphamide and doxorubicin with or with-
out cisplatin in advanced ovarian carcinoma: a Gyne-
cologic Oncology Group study. Cancer 1980; 57:1725-
1730.

4. Omura GA, Bundy BN, Berek JS, Curry S, Delgado G,

Mortel R. Randomized trial of cyclophosphamide plus
cisplatin with or without doxorubicin in ovarian carci-
noma: a Gynecologic Oncology Group study. J Clinic
Oncol 1989; 7:457^65.

5. Nejit JP, ten Bokkel Huinink WW, van der Burg MEL, et

al. Randomized trial comparing two combination che-
motherapy regimens (Hexa-CAF vs Chap-5) in ad-
vanced ovarian cancer. Lancet 1984; 2:594—600.

6. Schwartz PE, Smith JP. Second-look operations in ovari-

an cancer. Am J Obstet Gynecol 1980; 138:1124-1130.

7. Dougherty J, Hakes T, Caines J, Redman J. Recurrence

pattern of advanced ovarian carcinoma after negative
laparotomy. Proc Am Soc Clin Oncol 1985; 4:122.

8. Chambers SK, Chambers JT, Kohorn EI, Schwartz PE.

Etoposide (VP-16-213) plus cts-diamminedichloroplati-
num as salvage therapy in advanced epithelial ovarian
cancer. Gynecol Oncol 1987; 27:233-240.

9. Green JA, Slater AJ. A study of cis-platinum and ifosfa-

mide in alkylating agent-resistant ovarian cancer. Gy-
necol Oncol 1989; 32:233-240.

10. Trope C. Kaern J, Vergote I, Vossli S. A phase II study of

etoposide combined with ifosfamide as second-line ther-
apy in cisplatin-resistant ovarian carcinomas. Cancer
Chemother Pharmacol 1990; 26:845-847.

11. Dottino PR, Goodman HM, Kredentser D, Rosenberg M,

Cohen CJ. Clinical trial of etoposide and cisplatin as
salvage therapy in advanced ovarian carcinoma. Gyne-
col Oncol 1987; 27:350-356.

12. Evans BD, Raju KS, Calvert AH, Harland SSJ, Wiltshaw

E. Phase II study of JM8, a new platinum analog, in
advanced ovarian carcinoma. Cancer Treat Rep 1983;
67:997-1000.

13. Falkson G, Falkson HC. Further experience with isophos-

phamide. Cancer Treat Rep 1976; 60:955-957.

14. Hillcoat BL, Campbell JJ, Pepperell R, Quinn MA, Bishop

JF, Day A. Phase II trial of VP-16-213 in advanced
ovarian carcinoma. Gynecol Oncol 1985; 22:162-166.

15. Schabel FM Jr., Trader MW, Laster WR Jr., Corbett TH,

Griswold DP Jr. cjs-dichlorodiammineplatinum (II):
combination chemotherapy and cross-resistance studies
with tumors of mice. Cancer Treat Rep 1979; 63:1459-
1473.

16. Alberts DS, Canetta R, Masson-Liddil N. Carboplatin in

the first-line chemotherapy of ovarian cancer. Semin
Oncol 1990; 17(Suppl 2):54-60.

17. Yazigi R, Wild R, Madrid J, Arraztoa J. Ifosfamide treat-

ment of advanced ovarian cancer. Obstet Gynecol 1984;
63:163-166.

18. Markman M, Hakes T, Reichman B, et al. Ifosfamide and

ME8NA in previously treated advanced epithelial ovar-
ian cancer: activity in platinum-resistant disease. J
Clin Oncol 1992; 10:243-248.

19. Zalupski M, Baker LH. Ifosfamide. J Natl Cancer Inst

1988; 80:556-566.

20. Yang L, Rowe TC, Liu LF. Identification of DNA topo-

isomerase II as an intracellular target of antitumor epi-

316

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

podophyllotoxins in simian virus 40— infected monkey
cells. Cancer Res 1985; 45:5872-5876.

21. O'Dwyer PJ, Leyland-Jones B, Alonso MT, Marsoni S,

Wittes RE. Drug therapy: etoposide (VP-16-213). Cur-
rent status of an active anticancer drug. N Engl J Med
1985; 312(ll):692-700.

22. Schurig JE, Rose WC, Catino JJ, et al. The pharmacologic

characteristics of carboplatin: preclinical experience.
Carboplatin {JM-8). Current perspectives and future di-
rections. Philadelphia: WB Saunders, 1990:3-17.

23. McGuire WP, Rowinsky EK. Old drugs revisited, new

drugs, and experimental approaches in ovarian cancer
therapy. Semin Oncol 1991; 18:255-269.

24. Wiltshaw E, Kroner T. Phase II study of cis-dichlorodi-

ammineplatinum (II) (NSC-119875) in advanced adeno-
carcinoma of the ovary. Cancer Treat Rep 1976; 60:55-
60.

25. Bruckner HW, Cohen CJ, Wallach RC, et al. Treatment of

advanced ovarian cancer with cjs-diammineplatinum
II: poor-risk patients with intensive prior therapy. Can-
cer Treat Rep 1978; 62:555-558.

26. Gershenson DM, Kavanagh JJ, Copeland LJ, Stringer

CA, Morris M, Wharton JT. Re-treatment of patients

with recurrent epithelial ovarian cancer with cisplatin-
based chemotherapy. Obstet Gynecol 1989; 73:798-802.

27. Gore ME, Fryatt I, Wiltshaw E, Dawson T, Robinson BA,

Calvert AH. Cisplatin/carboplatin cross-resistance in
ovarian cancer. Br J Cancer 1989; 60:767-769.

28. Ozols RF, Ostchega Y, Curt G, Young RC. High-dose car-

boplatin in refractory ovarian cancer patients. J Clin
Oncol 1987; 5:197-201.

29. Colombo N, Speyer JL, Green M, et al. Phase II study of

carboplatin in recurrent ovarian cancer: severe hema-
tologic toxicity in previously treated patients. Cancer
Chemother Pharmacol 1989; 23:323-328.

30. Markman M, Hoskins W. Responses to salvage chemo-

therapy in ovarian cancer: a critical need for precise
definitions of the treated population (editorial). J Clinic
Oncol 1992; 10:513-514.

31. Meanwell CA, Blake AE, Latief TN, et al. Encephalopa-

thy associated with ifosphamide/MESNA therapy. Lan-
cet 1985; 1:406-407.

Submitted for publication August 1992.
Final revision received February 1993.

The Enteroinsular Axis and Endocrine
Pancreatic Function in Chronic
Alcohol Consumers:

Evidence for Early Beta-Cell Hypofunction

Renato J. Patto, M.D., Ewaldo K. Russo, M.D., Durval R. Borges, M.D., and Manoel M. Neves, M.D.

Abstract

Chronic alcohol consumers may have, as judged by functional criteria, exocrine as well as
endocrine pancreatic dysfunction, the latter represented by a decreased insulin response to
an oral glucose load. To investigate whether this decreased insulin response was due to an
ethanol-induced beta-cell dysfunction or to an ethanol-induced dysfunction of the entero-
insular axis, we determined glucose, insulin, and C-peptide plasma concentrations follow-
ing an oral and an intravenous glucose load in 16 healthy volunteer nonalcohol consumers
and in 10 chronic alcohol consumers. In each group, total integrated response for glucose
did not significantly change whether glucose was given orally or intravenously, indicating
isoglycemic glucose loads. The total integrated response values for insulin in the alcoholic
group following both glucose loads as well as C-peptide plasma concentrations were sig-
nificantly lower than in the control group. Moreover, in both groups the insulin TIR values
following the oral glucose load were significantly greater than the values obtained fol-
lowing the intravenous glucose load, indicating an incretin effect. These results indicate
that the decreased insulin response observed in alcoholics was not caused by a dysfunction
of the enteroinsular axis because it also occurred following an intravenous glucose load,
but by an ethanol-induced beta-cell dysfunction because C-peptide and insulin were pro-
portionally decreased in this group.

It is assumed that pancreatic hypersecretion of
bicarbonate and protein following secretin-chole-
cytokinin (S-CCK) stimulation is the first detect-
able functional alteration in asymptomatic alco-
holics (1). As we have previously reported (2),
many asymptomatic alcoholic subjects whose
S-CCK test showed normal secretion or hyperse-
cretion of bicarbonate had a decreased oral glu-
cose-induced insulin response that suggests that
the endocrine dysfunction might occur as early as

Partially supported by a grant to RJP (FMTM-83) from PICD/
CAPES-Brazil. From the Gastroenterology Sections of Facul-
dade de Medicina do Triangulo Mineiro-Uberaba, MG, and
Escola Paulista de Medicina — Sao Paulo, SP, Brazil. Address
correspondence and reprint requests to Dr. Manoel M. Neves,
Escola Paulista de Medicina, Rua Botucatu 862 Cx Postal
20239, 04023-Sao Paulo, SP-Brazil.

the exocrine dysfunction during the disease pro-
cess, although not always simultaneously (2).
Therefore, normal results in an S-CCK test on
those subjects does not exclude ethanol-induced
pancreatic injury.

It is well established that insulin secretion is
partially influenced by a complex mechanism, the
so-called enteroinsular axis (EIA), which com-
prises the action of gastrointestinal hormones (in-
cretin effect), adrenergic, cholingeric and pepti-
dergic neural transmission (3). Approximately
50% of the insulin released into the portal circu-
lation is degraded during its first circulation
through the liver, whereas C-peptide, which is
also released by beta cells in concentrations equi-
molar to insulin, is not (4). Therefore C-peptide
plasma concentrations better reflect the beta-cell
function than do plasma insulin concentrations.

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

317

318

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

TABLE

Glucose, Insulin, and C -Peptide Concentrations in Plasma
after an Oral and an Intravenous Glucose Load in Healthy
Controls and Alcoholics
(Mean ± SE)

Control Alcoholic

Factor in plasma

(n = 16)

(n = 10)

Glucose (mmol.L ')

Fasting

4.3 ± 0.2

4.1 ± 0.2

120-min post oral glucose

4.5 ± 0.4

4.2 ± 0.4

Glucose TIR

i mmrkl T . ^ 1 t^fi min ~ M

V IIIIIIUI* Li * l£t\J mill f

After oral load

613 ± 43

632 ± 42

After IV load

646 ± 39

642 ± 34

Insulin TIR

(nmoI.LM20 min"')

After oral load

43 ± 5

21 ± 2*

After IV load

28 ± 3t

12 ± l*t

Oral:IV ratio

1.6 ± 2.1

1.8 ± 1.0

C-peptide (nM)

After IV load at

corresponding insulin

peak rise

1.9 ± 0.2

1.1 ± O.lt

C-peptide:insulin ratio

2.4 ± 0.2

3.9 ± 1.2

TIR, total integrated responses.

* Significantly different ip < .05) from corresponding value for
oral test in both groups by Student's t test.

Significantly different (p < .005) from corresponding value
in controls by Student's t test.

t Significantly different (p < .05) from corresponding value in
controls by Student t test.

As ethanol can affect the pancreas (5) as well as
the liver and the gut (6), the decreased insulin
response observed in alcoholics could be a conse-
quence of (a) an increased hepatic insulin degra-
dation, (b) impaired EIA function, or (c) beta-cell
dysfunction, or of a combination of these states.

In the present study we determined glucose,
insulin, and C-peptide plasma concentrations fol-
lowing an oral and an intravenous (IV) glucose
loads in chronic alcohol consumers to examine the
EIA and beta-cell functions in these patients.

Materials and Methods

An oral (OGTT) and an intravenous (IVGTT)
glucose tolerance test were carried out in 15 male
and 1 female healthy volunteer nonalcohol con-
sumers (defined as occasional ethanol intake, al-
ways less than 50 g of pure ethanol), average age
35 (20-48) years, and in 10 chronic alcohol con-
sumers (called in this paper alcoholic subjects),
average age 43 (27-53) years, who had neither
clinical nor laboratory evidence of hepatic disease
or diabetes and have been consuming at least 50
g of pure ethanol daily for more than 2 years (2).
The average daily ethanol intake was 276 g (50-

480) for an average period of 21 (8-25) years. The
average total ethanol intake (daily intake x 365
X years) was 1,679.0 (548.0-2,920.0) kgs.

Measures of protein electrophoresis, pro-
thrombin time, bilirubin, ALT, AST, alkaline
phosphatase, and amylase, as well as plain ab-
dominal X-ray, were normal in all subjects. A diet
containing at least 100 g of carbohydrate was
given to all subjects during a 3-day period prior to
glucose tolerance tests which were performed on
subsequent days, the OGTT being the first. After
the subjects fasted overnight, blood samples were
collected for determination of fasting plasma glu-
cose levels. For the OGTT, a solution containing
100 g of glucose was given by mouth, and blood
samples were collected at 30, 60, 90, and 120 min-
utes. For the IVGTT, a 50% (w/v) glucose solution
(0.5 g/kg body weight) was administered by IV
bolus injection and blood collected at 1, 3, 5, 10,
20, 30, 40, 60, and 120 minutes (7). Plasma glu-
cose levels were immediately determined by an
oxidase method (God Papp, Boehringer Mann-
heim Biochemicals, Indianapolis, IN, US) and ab-
sorbances determined by a spectrophotometer
(505 nm). Plasma samples were stored at -70°C
for insulin (determined as previously described)
(2) and for C-peptide radioimmunoassay determi-
nations (kit from Immunonuclear Corporation,
Stillwater, MN, US). For each subject, C-peptide
concentration was determined in the plasma sam-
ples that corresponded to the insulin peak rise
following the IV glucose load. Total integrated
responses (TIR) for glucose and insulin were cal-
culated as described by Besterman et al. (8).

Statistical analyses were assessed by two-
way analysis of variance (ANOVA), and where
the ANOVA showed a significant difference, the
groups at each time point were compared using
the Student's t test. Informed consent was ob-
tained from each subject.

Results

Plasma glucose concentrations at fasting and
120-min after oral glucose load were normal in
both groups (Table). Moreover, in each group TIR
values for glucose following an oral glucose load
did not differ from those following an IV glucose
load, indicating isoglycemic glucose loads (Table).

In the control group, following the oral glu-
cose load, the plasma insulin concentration was
maximal at 30 min (fivefold increase) and de-
creased progressively to half-maximal at 120
min. In the alcoholic group there was a similar
pattern of increase, but with a significant de-
crease in the insulin curve (Fig. 1) and conse-
quently a decreased insulin TIR (Table).

Vol. 60 No. 4

ENTEROINSULAR AXIS AND ALCOHOLISM— PATTO ET AL.

319

In both groups, following the IV glucose load,
insulin concentrations rapidly increased to max-
imal at 5 min (sevenfold increase) and progres-
sively decreased to basal values at 120 min, but
the alcoholic group showed a distinct decrease in
the insulin curve (Fig. 2) and consequently de-
creased insulin TIR (Table). Consequently, the al-
coholic group showed significantly lower insulin:
glucose ratios in both tests (data not shown).

In each group, the insulin TIR value follow-
ing the oral glucose load was significantly greater
than that following the IV glucose load, indicat-
ing an incretin effect in both groups (Table).

The C-peptide mean value obtained for the
alcoholic group was significantly lower than that
obtained for the control group, whereas the
C-peptide:insulin ratio for the two groups was not
significantly different (Table).

Discussion

The present study illustrates that chronic al-
cohol consumers have a significantly decreased
insulin response following an oral glucose load,
confirming our previous observation (2), and il-

600 r
500

S 400

w 300

200
100

NGN ALCOHOLIC (CONTROL)
/

0 30 60 90 120
TIME (MINUTES)

Fig. L Plasma concentrations of insulin in response to oral
glucose load (100 g) in nonalcoholic (n = 16) and alcoholic (n
= 10) subjects. Values for plasma insulin concentrations were
determined in triplicate and results given are mean for the
number of subjects specified. *Significantly different (p < .05)
from corresponding control value by Student's t test.

D
C/)

NON ALCOHOLIC (CONTROL)

10 20 30

TIME (MINUTES)

Fig. 2. Plasma concentrations of insulin in response to in-
travenous glucose load (0.5 g/kg body weight) in nonalcoholic
(n = 16) and alcoholic (n = 10) subjects. Values for plasma
insulin concentrations were determined in triplicate and re-
sults given are mean for the number of subjects specified.
*Significantly different (p < .05) from corresponding control
values by Student's t test.

lustrates as well that the decreased insulin re-
sponse also occurs following an IV glucose load.

Our present results showed an incretin effect
in both groups and a decreased insulin response
to both oral and IV glucose loads in the alcoholic
group, making it unlikely that ethanol-induced
dysfunction of the EIA could account for the de-
creased insulin response observed in these pa-
tients. Normal EIA function has also been re-
ported in different situations, such as chronic
pancreatitis (9) and experimentally induced atro-
phy of the exocrine pancreas (10).

Because ethanol can affect the liver (6), it is
conceivable that ethanol could alter insulin up-
take by the liver and thus account for the de-
creased insulin response observed in these alco-
holics. However, a previous study (11) has
estimated the fractional hepatic uptake of insulin
from the molar ratio of insulin and C-peptide and
found that ethanol decreased the hepatic uptake
of insulin after an IV glucose load, which could
account for an eventual alcohol-induced in-
crease in insulin plasma concentrations but could
not account for an alcohol-induced decrease in in-

320

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

sulin plasma concentrations. Moreover, in alco-
holic patients with cirrhosis of the liver, glucose
intolerance and high plasma insulin concentra-
tions are usually found (12). In contrast, in the
present study we found a decreased insulin re-
sponse to a glucose load with normal glucose tol-
erance, since none of the control or the alcoholic
subjects met the National Diabetes Data Group
criteria (13) for either diabetes or impaired glu-
cose tolerance on the oral glucose tolerance test
performed. These findings, taken in conjunction
with those reported previously on the action of
ethanol on glucose uptake by the liver (9) and in
alcoholics with cirrhosis of the liver (12) argue
against the possibility of ethanol-induced liver
damage accounting for the decreased insulin re-
sponse observed in alcoholics.

The present finding of C-peptide and insulin
plasma levels proportionally decreased in alcohol-
ics indicates that the decreased insulin response
has occurred as a consequence of an alcohol-in-
duced beta-cell dysfunction. The mechanism by
which ethanol chronically ingested could affect
beta-cell function is not known. It has been re-
ported that alcohol induces degenerative changes
in human pancreatic acinar cells, suggesting a
direct toxic effect of ethanol (14). It has also been
proposed that ethanol chronically ingested in-
creases intrapancreatic cholinergic tone (15),
with consequent pancreatic dysfunction. How-
ever, no data is available to support either mech-
anism accounting for an alcohol-induced beta-cell
dysfunction.

In conclusion, the present results indicate
that chronic alcohol consumers may have de-
creased insulin response to oral and IV glucose as
a consequence of ethanol-induced beta-cell dys-
function. Therefore, glucose-induced insulin re-
sponse should be considered a complementary
procedure for the detection of ethanol-induced
pancreatic dysfunction in alcoholics, particularly
in those whose S-CCK test results are normal.

Acknowledgments

We thank Dr. Jerry D. Gardner for his suggestions and com-
ments during the preparation of the manuscript.

References

1. Neves MM, Borges DR, Vilela MP. Exocrine pancreatic

hypersecretion in Brazilian alcoholics. Am J Gastroent
1983; 78:513-516.

2. Patto RJ, Altikes I, Neves MM, Borges DR. Plasma insu-

lin levels in chronic alcoholics with and without exo-
crine pancreatic bicarbonate hypersecretion. Mt Sinai J
Med 1986; 53:558-562.

3. Creutzfeldt W, Ebert R. The enteroinsular axis. In: Go

VLN, Gardner JD, Brooks FP, Lebenthal E, Di-Magno
EP, Scheele GA, eds. The exocrine pancreas: biology,
pathobiology and diseases. New York: Raven Press,
1986, 313-346.

4. Hoekstra JBL, Van Rijn HJM, Erkelens DW, Thijssen

JHH. C-peptide: review. Diabetes Care 1982; 5:438-
446.

5. Singh M, Simsek H. Ethanol and the pancreas: current

status. Gastroenterology 1990; 98:1051-1062.

6. Burbige EJ, Lewis DR, Halsted CH. Alcohol and the gas-

trointestinal tract. Med Clin North Am 1984; 68:77-90.

7. Howanitz PJ, Howanitz JH. Carbohydrates. In: Henry JB,

ed. Clinical diagnosis and management by laboratory
methods, 17th ed. Philadelphia: WB Saunders, 1984,
172-174.

8. Besterman HS, Adrian TE, Bloom SR, et al. Pancreatic

and gastrointestinal hormones in chronic pancreatitis.
Digestion 1982; 24:195-208.

9. Stockmann F, Nauck M, Erhardt D, et al. In kretineffekt

bei Patienten mit exokriner Pankreasinsuffiziens in-
folge chronischer Pankreatitis. Aktuel EndoKrinol
1984; 5:125.

10. Folsch UR, Fussek M, Ebert R, Creutzfeldt W. Endocrine

pancreatic function during atrophy of the exocrine
gland. Pancreas 1988; 3:536-542.

11. Adner N. Influence of naloxone atropine and metoclo-

pramide on ethanol augmentation of insulin secretion
after intravenous glucose stimulation. Pancreas 1990;
5:460-466.

12. Shankar RP, Solomon SS, Duckworth WC, et al. Studies of

glucose intolerance in cirrhosis of the liver. J Lab Clin
Med 1983; 102:459^69.

13. National Diabetes Data Group — Classification and diag-

noses of diabetes mellitus and other categories of glu-
cose intolerance. Diabetes 1979; 28:1039-1057.

14. Noronha M, Salgadinho A, Ferreira de Almeida MJ,

Dreiling DA, Bordalo 0. Alcohol and the pancreas: I.
Clinical associations and histopathology of minimal
pancreatic inflammation. Am J Gastroenterol 1981; 76:
114-119.

15. Tiscornia OM, Celener D, Perec CJ, Lehmann ES, Cresta

MA, Dreiling DA. Physiopathogenic basis of alcoholic
pancreatitis: the effects of elevated cholinergic tone and
increased "pancreon" ecbolic response to CCK-PZ. Mt
Sinai J Med 1983; 50:369-387.

Submitted for publication September 1992.

Final revision received May 1993.

Thoracic Herniated Discs:

Review of the Literature and 12 Cases

Jeffrey S. Oppenheim, M.D., Allen S. Rothman, M.D., and Ved P. Sachdev, M.D.

Abstract

In comparison with herniations of lumbar or cervical intervertebral discs, symptomatic
thoracic disc herniation is rare. Between 1986 and 1991, 12 cases of thoracic herniated
discs were treated at The Mount Sinai Hospital, New York City. Most patients had back
pain or myelopathy. Nine of the disc herniations occurred at the lowest six thoracic in-
terspaces. Eight patients underwent costotransversectomy and discectomy. Seven of these
patients improved without complication. One patient was subsequently found to have a
spinal arteriovenous malformation below an incidental herniation that had been identi-
fied by magnetic resonance imaging. Because of the nonspecificity of the signs and symp-
toms, as well as the prevalence of incidental herniations on imaging, a careful clinical and
radiologic correlation is mandatory when diagnosing this uncommon pathology.

Although symptomatic herniation of a thoracic
intervertebral disc is extremely uncommon, it
merits attention for several important reasons.
First, the increased sensitivity provided by new
diagnostic techniques has heightened the need for
enhanced clinical specificity (1). Second, there
has been significant progress in surgical tech-
niques and approaches to the thoracic spine (2-
15). Finally, thoracic disc herniation is one of the
few notable causes of cord compression that is re-
versible.

According to Carson et al., the first descrip-
tion of a herniated thoracic disc occurred over 150
years ago (16). However, it was not until 1931
that Elsberg described the first surgically treated
patients (17). At that time these lesions were fre-
quently termed "eccondroses" or "ventral extra-
dural chondromas." So rare was their occurrence
that by 1960, the world literature included a total
of only 95 cases (17). The estimated annual inci-

Presented as a poster at the AANS Joint Section on Disorders
of the Spine and Peripheral Nerves, 8th Annual Meeting, Mi-
ami, FL, February 1992. From the Department of Neurosur-
gery, The Mount Sinai Medical Center, New York, New York.

Address reprint requests to Jeffrey S. Oppenheim, M.D.,
The Mount Sinai Medical Center, Box 1136, One Gustave L.
Levy Place, New York, NY 10029.

dence is one case of thoracic disc herniation per
one million population (16).

Materials and Methods

Twelve patients were treated in the Depart-
ments of Neurosurgery and Neurology of The
Mount Sinai Hospital in New York between Jan-
uary 1986 and June 1991 (Table 1). In addition to
a detailed neurologic examination, all patients
were evaluated by computed tomography (CT),
magnetic resonance imaging (MRI), or myelogra-
phy (Fig. 1). There were eight men and four
women, ranging in age from 33 to 72 years, with
a mean age of 53.

The most common presentation was back
pain or myelopathy. Myelopathy included pro-
gressive paraparesis, hyperreflexia, and a pin-
prick level. Two patients developed urinary re-
tention. The duration of symptoms ranged from
10 days to 8 years.

Nine of the herniated discs occurred at the
lower six thoracic interspaces (Fig. 1). Of the discs
occurring at the higher levels, two were at T5-6
and one was at T3-4. Only two cases were associ-
ated with prior trauma.

Eight patients were treated by costotransver-
sectomy and discectomy. This technique has been

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

321

322

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

Fig. 1. Thoracic myelogram shows epidural defect at T9-10
interspace causing posterior displacement of thecal sac.

well described elsewhere (15). Seven of these pa-
tients had relief of pain and significant improve-
ment of neurologic deficit and were able to walk
on discharge. There were no complications in
these seven patients. Three patients were man-
aged conservatively and were lost to follow-up;
one patient is considering surgery.

In one patient, who had rapidly progressive
paraparesis and myelopathy, MRI diagnosed a
T3-4 herniated disc. Despite costotransversec-
tomy and discectomy, this patient failed to im-
prove. A spinal arteriovenous malformation was
later discovered by myelogram three levels below
the herniated disc.

Discussion

Herniations of the intervertebral disc in the
thoracic spine are much less common than those

in the cervical or lumbar spine. An incidence of
between 0.2% and 5.3% of thoracic discs among
all herniated discs has been described, with a
mean of 0.67% (Table 2).

Thoracic disc herniation is predominantly a
disease of middle age. In a review of the litera-
ture, 80% of 288 patients were between the third
and fifth decades when initially consulting a phy-
sician, with a peak of 33% in the fourth decade
(18). The youngest patient was 12, the oldest 79.

The role of trauma has been debated as a
cause of herniated thoracic discs. Two of the larg-
est series found trauma to play an etiologic role in
about 25% of cases {n = 128). However, Russell
found that men under 40 years had a 53% inci-
dence of trauma whereas all others had a 17%
rate, suggesting that younger men represent a
distinct group in which trauma plays a signifi-
cant role (19).

Most series have found a male predominance
of herniated thoracic disc. Review of 10 series dis-
closed a 60% prevalence among men and a 40%
prevalence among women (Table 3), a finding
similar to that in this series: 67% men and 33%
women.

Most series find a predominance of central or
centrolateral herniated discs (Table 4). In a re-
view of 288 cases, the most common level was
Tll-12 (26%), and 75% of the thoracic discs oc-
curred below T8 (17). It has been suggested that
the more frequent localization in the lower tho-

TABLE 1

Clinical Features of 12 Herniated Thoracic Discs

Age/ Duration of Signs and HNP

# sex symptoms symptoms level Treatment Outcome

65M

3 wk

Back pain

Myelopathy

43F

Syr

Back pain

Myelopathy

51M

8 wk

Myelopathy

72M

6 wk

Back pain

Myelopathy

71M

10 d

Back pain

Myelopathy

60M

12 wk

Myelopathy

54M

4 wk

Back pain

Myelopathy

53F

6 wk

Back pain

Paresthesia

50F

8 wk

Back pain

Incont. x2d

50M

4 wk

Myelopathy

37F

2yr

Back pain

33M

Syr

Back pain

Myelopathy

T8-9

Costotransversectomy

Improved

T7-8

Conservative

Unchanged

T8-9
TlO-11

Costotransversectomy
Costotransversectomy

Improved
Improved

T7-8

Costotransversectomy

Improved

T3-4
TlO-11

Costotransversectomy
Costotransversectomy

Worsened
Improved

T5-6

Costotransversectomy

Improved

T6-7

Costotransversectomy

Improved

T5-6

T9-10

Tll-12

Conservative
Conservative
Conservative

LTF
LTF
LTF

LTF, lost to follow-up; F, female; M, male.

Vol. 60 No. 4

THORACIC HERNIATED DISCS— OPPENHEIM ET AL.

323

TABLE 2

Incidence of Thoracic Discs among All Discs

Discs

Series

inor

lotai

O^L 4-1-% AH

/o tnor

Mixter and Barr

1934 (34)

5.3

Love and Walsh

1938 (35)

113

5.3

Love and Kiefer

1950 (26)

5,500

0.2

Logue 1952 (27)

250

4.4

Abbott and Retter

1956 (36)

600

1.8

Arseni and Nash

1960 (17)

2544

0.47

Love and Schorn

1965 (20)

3683

0.46

Russell 1989 (19)

400

0.45

Total

13,109

0.67

Thor, thoracic.

racic spine is due to the increased mobility and
torsion stresses in this area (17, 18).

Difficulties in Diagnosis. Unlike lumbar and
cervical disc disease, thoracic disc herniation is
not associated with any distinct syndrome. As
might be expected, a central disc herniation is
likely to produce a myelopathy, whereas a lateral
herniation is apt to cause radicular symptoms. In
addition, the findings are dependent on the level
of the herniation. Further confounding the diag-
nosis is the nonspecificity of the most common
finding: back pain. This pain is variable in its
presentation, occurring locally or in a radicular
fashion. When it is radicular, a herniated thoracic
disc can be mistaken for cardiac, thoracic, or ab-
dominal pathology. Some patients have report-
edly been treated for gallbladder disease for a
misdiagnosed thoracic disc (18).

In Love and Schorn's series of 62 patients
(20), a correct preoperative diagnosis was made in
only 13 cases. In the era of MRI and CT, this low
preoperative diagnostic rate is unlikely, but it
does underscore the variability and nonspecificity
of the neurologic findings. After pain, the most
common symptoms are sensory disturbance and
motor weakness. The sensory abnormalities in-
clude both dysesthesias and paresthesias. Blad-
der involvement is a rare initial symptom but will
ultimately occur in one-third of patients (18).

In most series the duration of symptoms var-
ies from a few weeks to years. Histories as long as
26 years have been reported (18). The symptoms
can be constant or intermittent, or may progress
gradually or rapidly (21).

Plain X-rays can be helpful in the presence of
a calcified thoracic disc. In a review of 200 cases of
thoracic spine films, McAllister and Sage found
disc space calcifications in 4% (22). However, in
patients with herniated thoracic discs, calcifica-
tions are present in up to 75%. Therefore, the
presence of a calcified disc may suggest the pres-
ence of a herniation, but a normal plain X-ray of
the thoracic spine does not exclude a thoracic disc
protrusion.

Although myelography and computed tomog-
raphy have been the diagnostic procedure of
choice (Figs. 1, 2), MRI has recently been used
with greater frequency to evaluate patients with
herniated thoracic discs (Fig. 3). Several reports
have extolled the effectiveness of MRI in localiz-
ing symptomatic thoracic disc herniations (23,
24).

Despite high sensitivity for detecting tho-
racic discs, MRI may have poor specificity. One
study examined thoracic MR images obtained
from a group of 48 oncology patients who were not
known to have herniated discs (25). Using only

TABLE 3

Prevalence by Sex and Age of Thoracic Herniated Discs

Females Males Age range

Series No. (%) No. (%) (X = mean)

Love and Kiefer 1950 (26)

9 (53%)

8 (47%)

26-73 (X

= 48)

Kite et al. 1957 (28)

5 (50%)

18-68 (X

= 41.5)

Love and Schorn 1965 (20)

28 (46%)

33 (54%)

19-73

Benson and Byrnes 1975 (2)

7 (32%)

15 (68%)

33-66 (X

= 48)

Sekhar and Jannetta 1983 (14)

3 (25%)

9 (75%)

26-66 (X

= 50)

Maiman et al. 1984 (6)

6 (26%)

17 (73%)

16-69 (X

= 45)

Arce and Dohrman 1985 (18)

94 (40%)

142 (60%)

12-73 (X

= 45)

Bohlman and Zdeblick 1988 (3)

8 (42%)

11 (58%)

25-73 (X

= 36)

Otani et al. 1988 (9)

5 (22%)

18 (78%)

28-75 (X

= 48)

Russell 1989 (19)

31 (46%)

36 (54%)

16-79 (X

= 49)

Total

196 (40%)

294 (60%)

12-79

324

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

TABLE 4

Location of Disc Herniation

Central

or centro-

Series

lateral

Lateral

Love and Kiefer 1950 (26)

14 (77%)

3 (23%)

Perot and Munro 1969 (11)

57 (63%)

34 (37%)

Arce and Dohrman 1985 (18)

126 (68%)

60 (32% )

Tl-weighted images, these authors found a 14.5%
prevalence of incidental thoracic disc herniations.
This high prevalence of incidental herniations
underscores the need for a careful evaluation be-
fore making the diagnosis of symptomatic tho-
racic disc herniation. Given the lack of a distinct
clinical syndrome, confirmation of suspected her-
niations may require further study, including CT-
myelography in many patients.

Surgical Treatment. The surgical treatment
of thoracic disc disease involves problems unlike
the surgical treatment of cervical or lumbar disc
disease. For example, the thoracic cord has little
subarachnoid space and a tenuous blood supply.
In addition, displacement of the cord is limited by
the dentate ligaments. Centrally herniated tho-
racic discs are frequently hard and calcified and
are often adherent to the dura. Sometimes, dural
erosion occurs and a fragment of disc becomes em-
bedded in the spinal cord.

In response to these surgical challenges,
many approaches to the thoracic spine for the
treatment of herniated discs have been described,
including laminectomy, pediculectomy with joint
resection, the lateral extracavitary approach, the
lateral gutter approach, costotransversectomy,
and thoracotomy (2-16).

Fig. 2. Below Postmyelographic CT scan at T9-10 shows
calcified disc causing posterior displacement and disruption of
dye column. Fig. 3. Right Tl-weighted saggital (TR =
20, TE = 600) MRI reveals significant compression of spinal
cord by low signal mass at thoracic interspace level consistent
with calcified disc.

The results of most early surgical series are
disappointing. Until 1960, the standard treat-
ment consisted of decompressive laminectomy
with or without extradural or intradural removal
of the disc. Of the 17 cases of Love and Kiefer
reported in 1950, only 1 patient improved and
none completely recovered (26). In 1952, Logue
described 11 patients of whom only 6 improved
(27).

Kite et al. reported one case in 1957 of a phy-
sician who had a Tll-12 discectomy, presumably
by laminectomy (28). His paraparesis was unim-
proved and his pain worsened to the point that he
could no longer practice medicine and became ad-
dicted to narcotics. "A unilateral prefrontal lobot-
omy was performed . . . with little effect upon the
pain and was followed by the subsequent devel-
opment of postoperative epilepsy. Within 5 years
after removal of the disc, this physician under-
went complete dissolution of his personality and
died" (28). Despite the bad experience reported by
most series, laminectomy continues to be advo-
cated by some (1).

In 1971, Carson et al. described the lateral-
gutter approach, which combined laminectomy

Vol. 60 No. 4

THORACIC HERNIATED DISCS— OPPENHEIM ET AL.

325

TABLE 5

Rates of Surgical Success from Combined Series ( 1 7, 18)

Approach Total Cured Improved Unchanged Worse Died

Posterior 135 22(16%) 57(42%) 14(10%) 37(28%) 5(4%)
Lateral and

posterolateral 83 28(34%) 42(51%) 11(13%) 2(2%)

Transthoracic 52 22(42%) 25(48%) 2(4%) 3(6%)

with removal of a considerable portion of the fac-
ets (16). They noted that this allowed them to
approach the disc space very nearly laterally and
reported improvement in 12 of 14 cases. Patterson
and Arbitt reported a modified technique
whereby the patients were operated on by a mid-
line incision with removal of a facet and pedicle,
followed by discectomy and then laminectomy
(10). They described 3 patients; 2 were cured and
1 improved.

Costotransversectomy was first described by
Menard in 1900 for the surgical decompression of
tuberculosis (2). Hulme applied this technique to
thoracic disc herniations and reported improve-
ment in 4 of 6 patients operated on (29). Costo-
transversectomy provides a more lateral visual-
ization and therefore a greater opportunity for an-
terior maneuvering than does the lateral-gutter
technique. Although both interbody fusion (6) and
spinal instrumentation (30) have been advocated
following costotransversectomy to prevent instabil-
ity, long-term follow-up of patients without stabili-
zation finds no evidence of vertebral slippage (15).

The lateral approach is an extension of cos-
totransversectomy and involves a more lateral rib
resection. Although the approach was first de-
scribed by Capener, it was Larson who named it
the "lateral extracavitary approach" and reported
it in the treatment of traumatic lesions of the tho-
racolumbar spine (31). By 1984, Maiman et al.
reported 23 patients treated by this approach (6).
Seventeen had significant pain relief and 20 were
improved neurologically.

Crafoord et al. reported the first right-sided
transthoracic disc removal (32). The patient, who
had spastic paraparesis, made a complete recov-
ery. In 1969, Perot and Munro (11) and Ransohoff
et al. (12) simultaneously reported a total of 5 suc-
cessful anterolateral decompressions by thoracot-
omy. More recently, Otani et al. reported a series
of 23 patients in all of whom favorable outcomes
were achieved with a transthoracic extrapleural
approach (7). Others have found results compara-
ble to those of costotransversectomy (3, 33). How-
ever, some remain skeptical of this approach ow-
ing to the extensive nature of the surgery and the
potential for significant complications (30).

Conclusions

Because the manifestations of thoracic herni-
ated discs are protean, the diagnosis is often
highly dependent on radiographic studies. Given
a significant incidence of asymptomatic thoracic
herniations identified by MRI, great care must be
employed to avoid making a misdiagnosis (as oc-
curred in one case described here). It is therefore
recommended that myelography with CT be em-
ployed in addition to MRI in the following pa-
tients: those whose signs and symptoms do not
localize the lesion to a specific level; those in
whom the herniated disc occurs in the upper half
of the thoracic spine; and those with a coexistent
disease that might otherwise be responsible (e.g.,
metastatic carcinoma, neurofibromatosis).

Although few surgeons continue to advocate
a posterior surgical approach, the treatment of
herniated thoracic discs remains controversial.
And although many still prefer the various pos-
terolateral approaches, an increasing number of
surgeons are employing a transthoracic approach.
Some evidence suggests that the transthoracic
approach may offer better access to centrally her-
niated calcified discs (Table 5). At this time, the
decision on whether to use a posterolateral or an
anterolateral approach should depend on the
training and experience of the surgeon.

References

1. Singounas EG, Karvounis PC. Thoracic disc protrusion.

Acta Neurochir 1977; 39:251-258.

2. Benson MKB, Byrnes DP. The clinical syndromes and sur-

gical treatment of thoracic disc prolapse. J Bone Joint
Surg 1975; 57B;471^77.

3. Bohlman HH, Zdeblick TA. Anterior excision of herniated

thoracic discs. J Bone Joint Surg 1988; 7:1038-1047.

4. Dohn FD. Thoracic spinal cord compression: alternative

surgical approaches and basis of choice. Clin Neurosurg
1980; 27:611-623.

5. Lobosky JM, Hitchon PW, McDonnell DE. Transthoracic

anterolateral decompression for thoracic spine lesions.
Neurosurgery 1984; 14:26-30.

6. Maiman DJ, Larson SJ, Luck E, et al. Lateral extracavi-

tary approach to the spine for thoracic disc herniation:
report of 23 cases. Neurosurgery 1984; 14:178-182.

7. Otani K, Manzoki S, Shibasaki K, et al. Surgical treat-

ment of thoracic and thoracolumbar disc lesions using
the anterior approach. Spine 1977; 2:266-275.

8. Otani K, Nakai S, Fujimura Y, et al. Surgical treatment

326

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

of thoracic disc herniation using the anterior approach.
J Bone Joint Surg 1982; 64B:340-343.
9. Otani K, Yoshida M, et al. Thoracic disc herniation: Sur-
gical treatment in 23 patients. Spine 1988; 13(11):
1262-1267.

10. Patterson RH Jr., Arbitt E. A surgical approach through

the pedicle to protruded thoracic discs. J Neurosurg
1978; 48:768-772.

11. Perot PL, Munro DD. Transthoracic removal of midline

thoracic disc protrusions causing spinal cord compres-
sion. J Neurosurg 1969; 31:452^58.

12. Ransohoff J, Spencer F, Siew F, et al. Transthoracic re-

moval of thoracic disc: Report of three cases. J Neuro-
surg 1969; 31:459-461.

13. Safdari H, Baker RL. Microsurgical anatomy and related

techniques to an anterolateral transthoracic approach
to thoracic disc herniations. Surg Neurol 1985; 23:589-
593.

14. Sekhar LN, Jannetta PJ. Thoracic disc herniation: oper-

ative approaches and results. Neurosurgery 1983; 12:
303-305.

15. Young S, Karr G, O'Laoire SA. Spinal cord compression

due to thoracic disc herniation: Results of microsurgical
posterolateral costotransversectomy. Br J Neurosurg
1989; 3:31-38.

16. Carson J, Gumpert J, Jefferson A. Diagnosis and treat-

ment of thoracic intervertebral disc protrusions. J Neu-
rol Neurosurg Psychiatry 1971; 34:68-77.

17. Arseni C, Nash F. Thoracic intervertebral disc protrusion.

J Neurosurg 1960; 17:418-430.

18. Arce CA, Dohrman GJ. Herniated thoracic discs. Neurol

Clin 1985; 3:383-392.

19. Russell T. Thoracic intervertebral disc protrusion: expe-

rience of 67 cases and review of the literature. Br J
Neurosurg 1989; 3:153-160.

20. Love JG, Schorn VG. Thoracic disc protrusions. JAMA

1965; 191:627-631.

21. Reeves DL, Brown HA. Thoracic intervertebral disc pro-

trusion with spinal cord compression. J Neurosurg
1968; 28:24-28.

22. McAllister VL, Sage MR. The radiology of thoracic disc

protrusion. Clin Radiol 1976; 27:291-299.

23. Francavilla TL, Powers A, Dina T, Rizzoli HV. Case re-

port: MR imaging of thoracic disk herniations. JCAT
1987; 11:1062-1065.

24. Ross JS, Perez-Reyes N, Masaryk TJ, et al. Thoracic disk

herniation: MR imaging. Radiology 1987; 165:511-515.

25. Williams MP, Cherryman GR, Husband JE. Significance

of thoracic disc herniation demonstrated by MR imag-
ing. JCAT 1989; 13(2):211-214.

26. Love JG, Kiefer EJ. Root pain and paraplegia due to pro-

trusions of thoracic intervertebral discs. J Neurosurg
1950; 7:62-69.

27. Logue V. Thoracic intervertebral disc prolapse with spinal

cord compression. J Neurol Neurosurg Psychiatry 1952;
15:26-30.

28. Kite WC Jr, Whitfield RD, Campbell E. The thoracic her-

niated intervertebral disc syndrome. J Neurosurg 1957;
14:61-67.

29. Hulme A. The surgical approach to thoracic interverte-

bral disc protrusions. J Neurol Neurosurg Psychiatry
1960; 23:133-137.

30. Lesoin F, Rousseaux M, Autrique A, et al. Thoracic disc

herniations: evolution in the approach and indications.
Acta Neurochir 1986; 80:30-34.

31. Larson SJ, Hoist RA, Hemmy DC, Sances A Jr. Lateral

extracavitary approach to traumatic lesions of the tho-
racic and lumbar spine. J Neurosurg 1976; 45:628—637.

32. Crafoord C, Hiertonn T, Lindbloom K, Olsson SE. Spinal

cord compression caused by a protruded thoracic disc.
Report of a case treated with antero-lateral fenestration
of the disc. Acta Ortho Scand 1958; 28:103-107.

33. Albrand OW, Corkhill G. Thoracic disc herniation: treat-

ment and prognosis. Spine 1979; 4:41^6.

34. Mixter WJ, Barr JS. Rupture of the intervertebral disc

with involvement of the spinal canal. N Engl J Med
1934; 211:210-214.

35. Love JG, Walsh MN. Protruded intervertebral disks. Re-

port of one hundred cases in which operation was per-
formed. JAMA 1938; 111:396-100.

36. Abbott KH, Retter RH. Protrusions of thoracic interver-

tebral discs. Neurology 1956; 6:1-10.

Submitted for publication February 1992.

Screening for Human Immunodeficiency
Virus and Sexually Transmitted Diseases
in an Inner-City Colposcopy Clinic

Peter R. Dotting, M.D., Rhoda Sperling, M.D., and Romina Kee, M.D.

Abstract

Among patients attending an inner-city colposcopy clinic, the prevalence of Chlamydia
trachomatis was 22/375 (5.0%), Neisseria gonorrhoeae, 3/375 (0.8%), and seropositivity for
syphilis, 10/375 (2.7%). In addition, 13/261 (5.0%) of asymptomatic women agreeing to
voluntary human immunodeficiency virus (HIV-1) antibody screening were HIV-1 sero-
positive. Our data support incorporating screening for and education on sexually trans-
mitted diseases and HIV into the work of our colposcopy clinic.

Traditionally, colposcopy clinics have had a nar-
row focus as screening centers for cervical cancer.
The importance of a sexually transmitted patho-
gen, human papilloma virus (HPV), in the etiol-
ogy of cervical dysplasia and neoplasia has been
recognized; however, the precise role that HPV
plays in these disease processes remains unde-
fined (1-3).

The appreciation of the contribution of sexu-
ally transmitted HPV to cervical disease led us to
evaluate whether our colposcopy clinic should
routinely screen patients for the presence of other
sexually transmitted diseases (STDs), including
the human immunodeficiency virus (HIV-1).

Materials and Methods

The Mount Sinai Medical Center's colposcopy
clinic accepts referrals from the hospital-based
gynecology clinic and from other local health
agencies that serve the East Harlem community.
Approximately 300 new cases are referred yearly

From the Division of Gynecology Oncology (PRD) and Divi-
sion of Infectious Diseases (RS, RK), Department of Obstet-
rics, Gynecology, and Reproductive Science, The Mount Sinai
Medical Center, One Gustave Levy L. Levy Place, New York,
NY 10029. Address correspondence and reprint requests to
Peter R. Dottino, M.D., at that address.

for evaluation of Papanicolaou (Pap) smear ab-
normalities, including persistent inflammatory
atypia, evidence of human papilloma virus infec-
tion, cervical dysplasia, or cervical carcinoma.

From January 1, 1990, through May 1991, all
new clinic referrals received individual counsel-
ing concerning STD transmission, including HIV-
1, and the need for risk reduction behavior. A
questionnaire, completed by the interviewing au-
thor for each patient, included sociodemographic
information, a gynecologic and STD history, and
identification of high-risk behaviors for HIV-1 ac-
quisition. Patients accepting STD screening had
cervical samples assessed for Chlamydia tra-
chomatis by an antigen detection kit (Gen Probe,
San Diego, CA) and for Neisseria gonorrhoeae by
culture. A^. gonorrhea specimens were collected on
modified Thayer-Martin media (Becton Dickin-
son, Cockeysville, MD) and processed by standard
laboratory techniques. In addition, a repeat Pap
smear was performed on all patients. Serum sam-
ples for syphilis were tested by the rapid plasma
reagin (RPR) card test (Becton Dickinson, Cock-
eysville, MD), with confirmation of positive test
results by MHA-TP (Fujireba, Tokyo, Japan).
Those accepting voluntary screening for HIV-1
had antibody testing with a standard enzyme-
linked immunosorbent assay (ELISA); all posi-
tive results were confirmed by Western blot anal-
ysis.

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

327

328

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

TABLE 1

Results of Sexually Transmitted Disease Screening among
375 Women at an Inner-City Colposcopy Clinic

Positive results

Age

No. of

Chlamydia

Gonorrhea

Syphilis

(yr)

pts

No. {9c)

No. C/f)

No. {%)

13-19

8 (18)

1 (2)

0(0)

20-29

170

11 (6.5)

2(1.2)

8 (4.7)

30-39

100

2(2)

0(0)

1 (.27)

40-49

1 (3)

0(0)

0 (0)

50 +

0 (0)

0(0)

1 (.27)

Total

375

22 (5.9)

3 (0.8)

10 (2.7)

Results

A total of 375 patients were seen during the
study period; all agreed to gonorrhea, Chlamydia,
and syphilis screening, and 261/375 (70%) ac-
cepted HIV-1 antibody screening. Of those who
refused or were ineligible for HIV-1 antibody
screening, 5 were known to be HIV-1 seropositive,
1 initially refused testing but was subsequently
identified as seropositive, and 4 reported recent
negative antibody screens. Results of STD screens
are reported by age group in Table 1.

In the population screened, the prevalence of
C. trachomatis was 5.9% (95% CI 3.5%-8.3%), of
N. gonorrhoeae, 0.8% (95% CI 0%-1.7%), and of
positive results on serologic tests for syphilis,
2.7% (95% CI l%-4.3%). Of those who tested pos-
itive for sjrphilis, 6/10 (60%) had no prior history
of syphilis.

Of the 261 women agreeing to voluntary
HIV-1 antibody screening, 13/261 (5%) (95% CI
2.4%-7.6%) were identified as seropositive with
one additional test reported as inconclusive.
HIV-1 risk behaviors were identified by review-
ing answers from both the initial screening ques-
tionnaire and the follow-up posttest counseling

TABLE 2

A Comparison of Pap Smear Results and Cervical Biopsies
in 13 HIV-1 -Infected Women

no.

Pap smear

Biopsy result

Class III

CIN I

Class III

Inflammation

Class III

CIN I-II

Class III

CIN I

HPV

Koilocytosis

Class III

Koilocytosis

Class III

CIN II, CIN II on ECC

Class II

CIN I, koilocytosis

Class II

Inflammation

Class III

CIN III

HPV

Koilocytosis

Class III

CIN II

HPV

Inflammation

CIN, cervical intraepithelial neoplasia; HPV, human papilloma virus;
ECC, endocervical curettage.

interview. Of the newly identified HIV-1 seropos-
itive women, 5/13 (38%) had a history of injection
drug use (IDU), 4/13 (31%) had sexual partners
who were HIV-1 seropositive or had known high-
risk behaviors, and 4/13 (31%) had a history of
multiple sexual partners. Therefore, 62% appear
to have acquired their infection by heterosexual
contact. There were no differences in the propor-
tion of patients accepting screening among the
different age groups, and all newly identified
HIV-l-seropositive patients were between the
ages of 20 and 39. There were no positive screen-
ing tests for N. gonorrhoeae, C. trachomatis, or
Treponema pallidum among the newly identified
HIV-l-seropositive women.

Among the newly identified HIV-l-seroposi-
tive women, 11/13 underwent colposcopic exami-
nation and directed cervical biopsies. No cases of
cervical carcinoma were detected, and 7/11 pa-
tients had biopsies with a histologic appearance
consistent with HPV infection (Table 2). In the
HIV-1 -infected women, discrepancies of greater
than one grade in cytology and histology results
were not demonstrated.

Discussion

Patients in our colposcopy clinic demon-
strated a high rate of infections with STD patho-
gens. The infections identified do not necessarily
represent newly acquired infections but reflect
cumulative STD exposure of those patients. The
most frequent infection seen was C. trachomatis,
with rates of 18% and 6.5%, respectively, among
women 13-19 and 20-29 years of age. In other
populations studied, rates of infection have varied
from 3%— 5% in asymptomatic women to over 20%
in women in STD clinics (4—10). A recent survey
in Great Britain found that 23% of patients re-
ferred to a colposcopy clinic from both STD and
non-STD clinics were infected with C. trachoma-
tis and that the majority of patients were asymp-
tomatic (11). The small numbers of syphilis and
gonorrhea infections detected may reflect the
lower prevalence of these diseases in our commu-
nity and are consistent with rates noted in other
colposcopy clinics (11).

Among asymptomatic women accepting
HIV-1 antibody screening, 5% were identified as
seropositive; 8/13 (62%) were suspected of acquir-
ing their infection through heterosexual contact.
HIV-1 serosurveys of other populations of repro-
ductive-age women within our institution have
revealed a 16% seroprevalence rate in patients
admitted with pelvic inflammatory disease (12)
and a 2.7% seroprevalence rate in prenatal clinic

Vol. 60 No. 4

HIV AND STD IN COLPOSCOPY CLINIC— DOTTING ET AL.

329

patients (13). In New York City, a HIV-1 sero-
prevalence rate of 10% has been reported from
another inner-city colposcopy clinic (14).

Trends in the epidemiology of HIV-1 trans-
mission in women are disturbing. A review of ep-
idemiologic trends from an inner-city STD clin-
ic has revealed a twentyfold increase in HIV-1
seroprevalence during the past 10 years, the
sharpest increases reported being among Afri-
can-Americans, women, and adolescents (15). In-
creases in the ratio of female to male cases among
those reported with AIDS and the strong associ-
ation between the presence of genital ulcer dis-
ease, a common problem in sexually active
women, and HIV-1 acquisition further underscore
the risks to women (16-18).

The high prevalence of HIV and other STDs
in our colposcopy clinic population suggests that
all patients seen in our clinic should undergo STD
and HIV counseling and screening. This finding
may be generalizable to other comparable inner-
city colposcopy clinics, where comprehensive STD
and HIV education, screening, and treatment pro-
grams may help decrease the heterosexual spread
of these diseases. Monitoring trends in STD and
HIV prevalence in these clinics may provide im-
portant epidemiologic information about the het-
erosexual spread of these diseases.

References

1. Lancaster WD, Castellano C, Santos C, et al. Human pap-

illomavirus deoxyribonucleic acid in cervical carcinoma
from primary and metastatic sites. Am J Obstet Gyne-
col 1986; 154:115-119.

2. Lorincz AT, Temple GF, Kurman RJ, Jenson AB, Lan-

caster WD. Oncogenic association of specific human
papillomavirus types with cervical neoplasia. J Natl
Cancer Inst 1987; 79:671-767.

3. Reeves WC, Brinton LA, Garcia M, et al. Human papillo-

mavirus infection and cervical cancer in Latin America.
N Engl J Med 1989; 320:1437-1441.

4. Richmond SJ, Orio FT, DeBritton RC, et al. Value and

feasibility of screening women attending STD clinics
for cervical chlamydia infections. Br J Vener Dis 1980;
56:92-95.

5. McCormack WM, Alpert S, McComb DE, et al. Fifteen

month follow-up study of women infected with Chla-
mydia trachomatis. N Engl J Med 1979; 300:123-125.

6. Paavonen J, Saikku P, Vesterinen E, et al. Genital Chla-

mydia infections in patients attending a gynecological
outpatient clinic. Br J Vener Dis 1978; 54:257-261.

7. Ripa KT, Svensson L, Mardh PA, Westrom L. Chlamydia

trachomatis cervicitis in gynecologic outpatients. Ob-
stet Gynecol 1978; 52:698-702.

8. Oriel JD, Powis PA, Reever P, Miller A, Nicol CS. Chla-

mydial infections of the cervix. Br J Vener Dis 1974;
50:11-16.

9. Saltz GR, Linnemann CC, Brookman RR, Rauh JL. Chla-

mydia trachomatis cervical infection in female adoles-
cents. J Pediatr 1981; 98:981-985.

10. Bowie WR, Borrie-Hume CJ, Manzon LM, et al. Preva-

lence of C. trachomatis and N. gonorrhoeae in two pop-
ulations of women. Can Med Assoc J 1981; 124:1477-
1479.

11. Byrne MA, Turner MJ, Griffiths M, Taylor-Robinson D,

South WA. Evidence that patients presenting with dys-
karyotic cervical smears should be screened for genital-
tract infections other than human papillomavirus in-
fection. Eur J Obstet Gynecol Reproduct Biol 1991; 41:
129-133.

12. Sperling RS, Friedman F, Joyner M, Brodman M, Dottino

P. Human immunodeficiency virus seroprevalence in
hospital admissions with pelvic inflammatory disease.
J Reproduct Med 1991; 36(2):122-124.

13. Sperling RS, Sacks HS, Mayer L, Joyner M, Berkowitz R.

Umbilical cord blood serosurvey for human immunode-
ficiency virus (HIV) in parturient women in a voluntary
hospital in NYC. Obstet Gynecol 1989; 73:179-181.

14. Maiman M, Fruchter RG, Serur R, Boyce JG. Prevalence

of human immunodeficiency virus in a colposcopy clinic
(letter). JAMA 1988; 260:2214.

15. Quinn TC, Groseclose SL, Spence M, Provost V, Hook EW

III. Evolution of the human immunodeficiency virus ep-
idemic among patients attending sexually transmitted
disease clinics: a decade of experience. J Infect Dis
1992; 165:541-544.

16. Moss CG, Kreiss JK. The interrelationship between hu-

man immunodeficiency virus infection and other sexu-
ally transmitted diseases. Med Clin North Am 1990;
74:1647-1660.

17. Plummer FA, Simonsen IN, Cameron DW, et al. Cofactors

in male-female sexual transmission of the human im-
munodeficiency virus type 1. J Infect Dis 1991; 163:
233-239.

18. Hook EW III, Cannon RO, Nahmias AJ, et al. Herpes

simplex virus infection as a risk factor for human im-
munodeficiency virus infection in heterosexuals. J In-
fect Dis 1992; 165:25-35.

Submitted for publication August 1992.
Final revision received April 1993.

Case Report:

Evolution of a Type B Aortic Dissection following Renal

Artery Angioplasty

Robert Gendler, M.D., and Harold A. Mitty, M.D.
Abstract

Percutaneous renal artery angioplasty is an accepted treatment for renal artery stenosis.
A variety of complications have been reported related to this procedure. The authors
report a case of Type B aortic dissection occurring immediately following renal artery
angioplasty, a previously unreported complication of renal artery angioplasty. A mecha-
nism for this complication in this patient is postulated.

During the past 10 years percutaneous angio-
plasty has become the standard of treatment for
renovascular hypertension caused by renal artery
stenosis. Reported complications include occlu-
sion of the renal artery (secondary to thrombosis,
dissection, or embolization), perforation of the re-
nal artery (that is, with guide wire), renal artery
rupture, puncture site hematomas, and contrast-
induced renal failure (1-3). To our knowledge
there is no report of type B aortic dissection de-
veloping in relation to renal artery angioplasty
(1-4). We report a case of a type B aortic dissec-
tion that evolved immediately following percuta-
neous angioplasty of a left renal artery stenosis.

Case Report

A 56-year-old woman with a 20-year history
of hypertension was admitted to our hospital with
crushing chest pain radiating to her back. The
patient's hypertensive medication included Cloni-
dine 0.3 mg every day. Propranolol 20 mg three
times a day, Cardizem 30 mg four times a day,
and Isordil 20 mg three times a day. The patient's
blood pressure on admission was 220/130 mm Hg

From the Department of Radiology, Mount Sinai Medical Cen-
ter, One Gustave L. Levy Place, Box 1234, New York, New
York 10029. Address reprint requests to Harold A. Mitty,
M.D., at that address.

and she admitted not following her medication
prescriptions recently. Although the patient had
a long history of stable angina controlled by med- I
ication, these symptoms were atypical for her, |
and the possibility of aortic dissection was sus- ^
pected. EKG showed normal sinus rhythm at 91
BPM, evidence of left ventricular hypertrophy,
and nonspecific T-wave changes, without acute
ischemic changes. Hematocrit was 31.5%, BUN
126 mg/dL, creatinine 6.2 mg/dL. Urine analysis
showed pH 5, protein 30 mg/dL, glucose 50 mg/dL,
red blood cell count 0-2/HPF, whole blood cells
1/HPF.

The patient underwent transesophageal t
echocardiography, which showed left ventricular f
hypertrophy, mild aortic root dilatation, but no
evidence of aortic dissection (Fig. 1). The patient
was admitted to the intensive care unit and her
blood pressure was brought under control by
means of a nitroprusside drip. The patient's chest
pain abated. Her creatinine level, which was nor- !
mal 2 months prior to admission, was now 6.2
mg/dL. A Doppler sonogram of the kidneys
showed a normal-sized right kidney (10.2 cm) and
normal right renal artery. The left kidney was
small (7.4 cm) and there was high velocity flow in
the left renal artery, suggesting stenosis. Nuclear j
renal scan showed diminished flow bilaterally, i'
which could be due to bilateral renal artery ste- j
nosis. The patient was sent for angiography. p

330

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

Fig. 1. Transesophageal echocardiogram at level of descending thoracic aorta
showing no evidence of dissection.

Fig. 2. Left Abdominal aortogram performed prior to angioplasty showing significant left renal artery stenosis and mild right renal
artery stenosis. There is no evidence of aortic dissection. Fig. 3. Right Aortic flush performed immediately after angioplasty
demonstrating false lumen above and below left renal artery (arrows).

Fig. 4. Representative image from CT scan performed ap-
proximately 1 hour after angioplasty showing aortic dissec-
tion.

332

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

Aortography demonstrated high-grade steno-
sis of the proximal right renal artery (Fig. 2).
There was no gradient across the right-sided le-
sion. It was decided to dilate the left renal artery
stenosis in the hope of improving the patient's
hypertension and renal function.

Upon dilating the left renal artery stenosis
with a 5-mm-diameter, 2-cm length balloon, the
patient began to experience severe back pain. A
postangioplasty study was performed with a flush
catheter near the left renal artery ostium. The
digital arteriogram (Fig. 3) showed narrowing of
the infrarenal aortic lumen and filling of a false
lumen seen along the left renal artery. A commu-
nication (entry point) was seen just superior to
the ostium of the newly dilated left renal artery.
The patient's blood pressure was 160/100 mm Hg.
No pressure gradient remained across the left re-
nal artery.

Because of the findings on the postangio-
plasty aortogram and the patient's symptoms, she
was sent immediately for a dynamic CT scan
(Fig. 4) of the chest and abdomen. The CT scan
showed a type B aortic dissection starting at the
left subclavian artery and extending to the ab-
dominal aorta distal to the left renal artery. The
patient was treated conservatively in the inten-
sive care unit with blood pressure control. The
patient's renal function failed to improve, and she
was placed on long-term hemodialysis. Her blood
pressure has been brought under control on oral
medications and repeat CT scan showed the aortic
dissection to be stable 6 months after the proce-
dure.

Discussion

Aortic dissection is not a reported complica-
tion of renal artery angioplasty (1-4). In a recent
review of aortic dissection by Cigarroa et al, renal"
artery angioplasty was not listed as a cause of
aortic dissection (5). In our patient, evolution of
the aortic dissection clearly was related to dilata-
tion of the renal artery. The dissection was not
related to guide-wire or catheter manipulation,
since we have evidence on a predilation angio-
gram that the aortic caliber was normal. In addi-
tion, the temporal relationship of the patient's in-
creased back pain after the dilation points to the
dilation as an exacerbating event.

This patient had symptoms suggesting aortic
dissection prior to the procedure. This led to a
transesophageal echocardiogram in the emer-
gency room. This procedure is reported to have a
sensitivity of 99% in the detection of aortic dis-

section (6). The patient may have had medial ne-
crosis of her aorta and was possibly evolving an
aortic dissection at the time of that initial episode
of back pain. Once the diagnosis of aortic dissec-
tion was excluded by the echocardiogram, clinical
attention centered on her hypertension. The pa-
tient apparently had accelerated hypertension
aggravated by noncompliance with medications.
The accelerated hypertension and deterioration
in her renal function, along with the ultrasound
and nuclear medicine studies, raised the suspi-
cion of renal artery stenosis. The patient then un-
derwent the angioplasty, which most likely accel-
erated the aortic dissection by creating an entry
(or reentry) point into a diseased media.

Retrograde type B aortic dissections are rare.
Even guide-wire retrograde dissection is easily
recognized and rarely extends. It is unlikely that
the angioplasty was the sole initiating event in
our patient. It is more probable that the dilata-
tion served as an exacerbating event by creating
a reentry point into an already diseased media.

Summary

The partial disruption of the arterial wall
that occurs with angioplasty may produce an en-
try point into a vessel with preexisting cystic me-
dial necrosis. At present, there is no clinically re-
liable method of detecting medial necrosis. One
should proceed with caution if angioplasty is in-
dicated and dissection has been recently consid-
ered as a clinical possibility.

References

1. Sos TA, Pickering TG, Saddekni S, et al. The current role

of renal artery angioplasty in the treatment of renovas-
cular hypertension. Urol Clin North Am 1984; (3) SOS-
SIS.

2. Martin LG, Casarella WJ, Alspaugh JP, et al. Renal ar-

tery angioplasty: increased technical success and de-
creased complications in the second 100 patients. Radi-
ology 1986; 159:631-634.

3. Tegtmeyer CJ, Kellum CD, Ayers C. Percutaneous trans-

luminal angioplasty of the renal arteries. Radiology
1984; lS3:77-84.

4. Tegtmeyer CJ, Selby JB Jr. Percutaneous transluminal

angioplasty of the renal arteries. In: Castaneda-Zuniga
WR, Tadavarthy SM, eds. Interventional radiology, 2nd
ed. Baltimore: Williams & Wilkins, 1992, 364-378.

5. Cigarroa JE, Isselbacher EM, DeSanctis RW, Eagle KA.

Diagnostic imaging in the evaluation of suspected aor-
tic dissection. N Engl J Med 1993; 328(1 ):35^3.

6. Erbel R, Daniel W, Visser C, et al. Echocardiography in

the diagnosis of aortic dissection. Lancet 1989; 4:4S7-
460.

Submitted for publication January 1993.
Final revision received June 1993.

Propofol Use in
Malignant Hyperthermia:

A Case Report

Gordon Freedman, M.D., Ian H. Sampson, M.D., and Steven Cagen, M.D.

Abstract

We report a case of a patient who developed unexpected intraoperative malignant hyper-
thermia terminated after discontinuation of volatile agents and kept anesthetized with
propofol until the completion of the operation.

Malignant hyperthermia may be triggered by
many drugs during anesthesia, including potent
inhalational agents and succinylcholine. Propofol
has been used successfully in patients with both
suspected and proven susceptibility to malignant
hyperthermia (1-3). In vitro studies with animals
and humans have shown it not to be a triggering
agent (4). We report a case of the successful ter-
mination of a malignant hyperthermia reaction
after discontinuation of volatile agents and con-
version to propofol anesthesia.

Case Report

A previously healthy 32-year-old, 110 kg wo-
man was scheduled for left partial thyroidectomy
for a slowly enlarging thyroid mass. Preoperative
thyroid function tests showed the patient to be
euthyroid. She had previously received an uncom-
plicated general anesthetic for cervical dilatation
and curettage, but information regarding exactly
which anesthetics were used was unobtainable.
Her mother had undergone general anesthesia for
a craniotomy, also without complications.

One hour before surgery the patient was se-

From the Department of Anesthesiology, The Mount Sinai
Hospital and Mount Sinai School of Medicine, New York, NY.
Address reprint requests to Gordon Freedman, M.D., Depart-
ment of Anesthesiology, Box 1010, Mount Sinai Medical Cen-
ter, One Gustave L. Levy Place, New York, NY 10029-6574.

dated with oral diazepam 10 mg. Following a de-
fasciculation dose of curare 3 mg, anesthesia was
induced with fentanyl 2 mcg/kg, thiamylal 5 mg/
kg, and succinylcholine 1.5 mg/kg. After unevent-
ful tracheal intubation, anesthesia was main-
tained with N2O 70%, O2 30%, isoflurane 0-1%,
fentanyl, and pancuronium. Frozen-section bi-
opsy revealed the mass to be a well-differentiated
papillary carcinoma and lymph node dissection
was to be included in the procedure.

Approximately 45 minutes into the operation
the surgeon noticed tightening of the strap mus-
cles of the neck. Several additional doses of pan-
curonium failed to facilitate relaxation. Over the
next 30 minutes the heart rate increased from 95
to 125 beats per minute; the end tidal CO2
(ETCO2) increased from 29 to 60 mmHg despite
doubling the minute ventilation to 22,000 L/min,
and the esophageal temperature increased from
36.9° to 38.4° C. At this time the end tidal con-
centration of isoflurane was 1.1%. An arterial
blood gas examination showed pH 7.24, PaC02 57
mmHg, Pa02 128 mmHg, and base excess -3.4
mEq/L. The surgeon was informed of these
changes and asked to finish the operation as soon
as possible. The inhalational anesthetics were
turned off and 100% O2 administered. Fentanyl
2.5 mcg/kg and propofol at 0.1 mg/kg per minute
were given in intermittent boluses to maintain
anesthesia, and the operation was concluded in
approximately 45 minutes. With the removal of

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

333

334

THE MOUNT SINAI JOURNAL OF MEDICINE

September 1993

the inhalational agents, ETCO2 decreased to 34
mmHg and both heart rate and temperature de-
creased over the next 10 minutes to their previous
levels. By this time solubilized dantrolene had
been made ready, but because of the marked clin-
ical improvement it was decided to keep it in re-
serve.

After uneventful emergence, she was trans-
ferred to the recovery room, where a urine speci-
men was found to be positive for myoglobin and a
blood specimen positive for free serum myoglobin.
Serial creatine phosphokinase (CPK) determina-
tions were obtained at 6, 12, and 24 hours. The
preoperative CPK had been 307 lU; the initial
reading in the recovery room of 6,354 lU peaked
to 35,000 lU six hours later and 39,000 at 24
hours, then decreased gradually. Over the next
three days the serum calcium levels increased
and serum phosphate levels decreased, consistent
with rhabdomyolysis. Treatment was instituted
with forced saline diuresis, maintaining urine
output over 150 mL per hour, and urinary alka-
linization was achieved with intravenous sodium
bicarbonate to keep the urine pH over 6.5. Blood
urea nitrogen and creatinine levels remained nor-
mal. Neither the patient nor her family gave any
history of myopathy. On the basis of these clinical
findings and the blood test results, a diagnosis of
malignant hyperthermia was made.

Discussion

Diagnosis of Malignant Hyperthermia. Ma-
lignant hyperthermia was first described in 1960
by Denborough and Lovell (5). It is a heritable
disorder of skeletal muscle initiated by several
triggering agents, resulting in elevated myoplas-
mic calcium levels. The exact pathophysiology
causing malignant hyperthermia is not yet well
defined. It is thought to involve an alteration in
the function of several proteins in skeletal muscle
secondary to elevated levels of certain fatty acids
or their metabolites. The chemical properties of
the triggering agents that would cause these al-
terations has not been determined.

Clinical signs of malignant hyperthermia in-
clude tachycardia, tachypnea, increased blood
pressure, muscle rigidity, and hyperthermia.
Laboratory abnormalities include hypercarbia,
hyperkalemia, respiratory and metabolic acido-
sis, hypercalcemia, myoglobinuria, and elevated
CPK. Malignant hyperthermia is rare, occurring
in approximately 1 in 50,000 anesthetized adults.

Clinically, this patient fits the picture of ma-
lignant hyperthermia. She was exposed to the
triggering agents, succinylcholine and isoflurane,

then intraoperatively developed marked muscle
rigidity, tachycardia, hypercarbia, marked respi-
ratory acidosis, and mild metabolic acidosis and
hyperthermia. Subsequent laboratory tests re-
vealed a CPK greater than 39,000 lU and myo-
globinuria. In patients with masseter muscle ri-
gidity and no history of myopathy, a CPK value
greater than 20,000 lU is a highly suggestive in-
dicator of malignant hyperthermia susceptibility
(6).

Dantrolene was not given to this patient be-
cause of the rapid normalization of vital parame-
ters on discontinuation of the triggering agents.
Dantrolene does have side effects such as muscle
weakness, nausea, and phlebitis and since the pa-
tient's clinical picture was stabilizing, it was de-
cided to closely observe the patient in the recov-
ery room, keeping dantrolene ready if needed.
Consultation with the malignant hyperthermia
hotline recommended giving dantrolene only if
signs of recrudescence occurred, in the form of
worsening clinical situation (blood pressure,
heart rate, temperature, electrolytes, arterial
blood gases) or an increase in CPK levels after the
initial peak (in this case the CPK peaked at
39,000 lU and gradually decreased to normal lev-
els).

Thyroid storm was ruled out as a diagnosis
by the presence of muscle rigidity, an elevated
CPK, and the euthyroid preoperative state of the
patient. In a previous report, two cases of intra-
operative hyperthyroidism were misdiagnosed as
malignant hyperthermia (7). However, in both of
these patients the thyroid function tests were sig-
nificantly elevated.

On several occasions since this event both the
patient and her family have been strongly urged
to undergo a muscle biopsy for halothane-caffeine
testing (patients with malignant hyperthermia
have an exaggerated muscle contracture response
to these substances), but have not done so to date.
This test is important to make a definitive diag-
nosis of this heritable disease for this patient and
her family to prevent any future anesthetic mis-
haps.

Use of Propofol. Propofol, chemically 2,6-
diisopropylphenol, is an intravenous hypnotic
agent for induction and maintenance of anesthe-
sia. In vitro studies in which muscle samples from
two people and two pigs susceptible to malignant
hyperthermia were exposed to propofol, halo-
thane, and caffeine have demonstrated that both
halothane and caffeine caused contracture but
propofol did not (4). Propofol has been shown not
to precipitate malignant hyperthermia in geneti-
cally susceptible swine (8, 9). It has also been

Vol. 60 No. 4

PROPOFOL AND MALIGNANT HYPERTHERMIA— FREEDMAN ET AL

335

used successfully in humans with medical histo-
ries suggesting potential for malignant hyper-
thermia (1) and in humans with biopsy-proven
malignant hyperthermia (2, 3). The biochemical
basis for propofol's nonpredilection to cause ma-
lignant hyperthermia has not been elucidated. To
our knowledge, this is the first reported case of
the use of propofol after the onset of an acute ma-
lignant hyperthermia reaction in humans.

Propofol enabled us to stop administering
N2O and to administer 100% O2 at a time when
the patient's oxygen consumption was greatly el-
evated. In addition to fentanyl for analgesia,
propofol facilitated adequate anesthesia, allowing
the surgeon to quickly terminate the operation
without risk of patient awareness and without ex-
acerbating the acute malignant hyperthermia ep-
isode.

Summary

This case illustrates that conversion to propo-
fol anesthesia during an acute episode of malig-
nant hyperthermia may facilitate anesthesia
during termination of this potentially lethal com-
plication.

References

1. Richardson J. Propofol infusion for coronary artery bypass

surgery in a patient with suspected malignant hyper-
pyrexia. Anae-sthesia 1987; 42:1125.

2. Cartwright DP. Propofol in patients susceptible to malig-

nant hyperpyrexia. Anaesthesia 1989; 44:173.

3. Marks LF, Edwards JC, Linter SPK. Propofol during car-

diopulmonary bypass in a patient susceptible to malig-
nant hyperpyrexia. Anaesth Intensive Care 1988; 16:
482-485.

4. Denborough M, Hopkinson KC. Propofol and malignant

hyperpyrexia. Lancet 1988; 1:191.

5. Denborough MA, Lovell RRH. Anaesthetic deaths in a

family. Lancet 1960; 2:45.

6. Rosenberg H, Fletcher JE. Masseter muscle rigidity and

malignant hyperthermia susceptibility. Anesth Analg
1986; 65:161-164.

7. Peters KR, Nance P, Wingard DW. Malignant hyperthy-

roidism or malignant hyperthermia? Anesth Analg
1981; 60:613-615.

8. Raff M, Harrison GG. The screening of propofol in MHS

swine. Anesth Analg 1989; 58:750-751.

9. Krivosic-Horber R, Reyfort H, Becq MC, Adnet P. Effect of

propofol on the malignant hyperthermia susceptible pig
model. Br J Anaesth 1989; 62:691-693.

Submitted for publication November 1992.
Final revision received June 1993.

The

Formulary

Isoniazid-Induced Hepatotoxicity

During the past several months, The Mount
Sinai Hospital Adverse Drug Reaction Subcom-
mittee has received three reports of isoniazid-in-
duced fulminant hepatic failure, requiring liver
transplantation.

The resurgence of tuberculosis has resulted
in an increase in the use of isoniazid (INH), along
with the other antitubercular medications. Al-
though the incidence of INH-resistant M. tuber-
culosis is reported to be as high as 40% in New
York City, INH prophylaxis and treatment are
still frequently utilized and it is important that
practitioners remain aware of the risks of hepa-
totoxicity associated with its use.

The introduction of INH in 1952 was a major
breakthrough in the treatment of tuberculosis be-
cause of its high effectiveness. A few cases of
drug-induced jaundice were reported soon after
its introduction; however, causality was often at-
tributed to concurrent viral hepatitis or the con-
comitant use of other hepatotoxic agents.

Since the introduction of INH, clinical hepa-
titis and elevation of liver function test results in
asymptomatic patients taking INH have been re-
ported. About 10% to 20% of patients receiving
this drug develop transient mild elevations of
SGOT that usually disappear despite continued
therapy. These patients should be closely moni-
tored for the possible progression of hepatic ne-
crosis, and therapy should be discontinued in
those patients whose transaminase levels exceed
three times the normal values. An advisory com-
mittee appointed by the Centers for Disease Con-
trol concluded that hepatitis occurs in approxi-
mately 1 to 10 patients per thousand treated with
INH. Fatal hepatotoxicity can occur in approxi-
mately 10% of these patients.

The mechanism of INH hepatotoxicity is not
completely understood and is felt to resemble a
hypersensitivity reaction. Isoniazid hepatitis is
characterized by histologic lesions of hepatocellu-
lar injury, along with inflammation and necrosis
which is clinically indistinguishable from that oc-
curring in viral hepatitis. Although in some cases

the presence of eosinophils is prominent, the in-
flammatory response is primarily mononuclear.

Preventive use of INH has generated some
controversy since 1975, although official agencies
continue to advocate its use. The following per-
sons are considered to be candidates for preven-
tive therapy:

• those who are PPD positive (5 mm) and infected
with HIV, or those at risk for HIV infection
who decline HIV testing

• those in close contact (PPD 5 mm) with persons
newly diagnosed as having infectious tubercu-
losis

• recent converters, as indicated by a Mantoux
test with an increase 10 mm or greater within
a two-year period

• persons with an abnormal chest x-ray that
shows fibrotic lesions likely to represent old
healed tuberculosis (5 mm)

• persons with medical conditions known to in-
crease the risk of active tuberculosis if infection
has occurred (10 mm)

• all PPD-positive (10 mm) persons under 35
years of age

Practitioners may want to consider the risk-ben-
efit of using INH in:

• those patients with suspected drug-resistant
tuberculosis

• those at significant risk of INH hepatotoxicity

Risk factors include age greater than 35 years
and chronic alcohol use.

Baseline liver function tests, monitoring of
liver function during therapy, and approriate pa-
tient counseling is recommended for all patients
receiving INH, since hepatotoxicity can develop
early or late in the course of therapy. The pro-
drome to jaundice has been characterized by
weakness, anorexia, fatigue, diarrhea, and mal-
aise. It is important to inform patients about and
monitor them for these symptoms, and to discon-
tinue INH use immediately if these symptoms ap-
pear in order to prevent serious hepatotoxicity.

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

337

Mount Sinai School of Medicine
Awards and Prizes
Academic Year 1992/93

Arthur H. Aufses, Sr., Prize in Surgery
Robert Jioven Valenzuela

Basic Sciences Aciiievement Award
Laura Ann Karch

Dr. Morris B. Bender Family Award in Clinical Neurology
John Attilio Martignetti, Ph.D.

Dr. Morris B. Bender Family Award in the Neurosciences
Todd Rosenzweig
(Class of 1995)
Amit Schwartz
(Class of 1995)

Cecilia Yoon
(Class of 1995)

Doctoral Dissertation Awards
Delong Liu, M.D., Ph.D.
(Graduate School of Biological Sciences)

Bo Yu, Ph.D.
(Graduate School of Biological Sciences)

Harold Elster Memorial Prize for Highest
Academic Standing
Lorraine M. Lazar, Ph.D.

James Felt Memorial Prize for Highest
Overall Performance
Erika Suzanne Berman

Dr. Alan F. Guttmacher Obstetrics & Gynecology Prize
Karen Lynn Patrusky

The Milton Handler Award
Amy B. Eisenberg
Bernice Desiree Griffith

Saul Horowitz, Jr., Memorial Award
Scott L. Friedman, M.D.

(Class of 1979)
Charles M. Miller. M.D.

Dr.

Dr. M. Ralph Kaufman Psychiatry Prize
Stephanie Marie Le Melle
Helene M. McDonald

Dr. Harold Lamport Biomedical Research Prize
Eugene Toy
(Class of 1995)

Medical Attendings Clinical Medicine Award
Adina Holand
Biyan Christopher Yen

Medical Society of the State of New York
Award for Outstanding Community Service
John Ladislav Bucek

Mount Sinai Auxiliary Board Prize for Excellence in
Study of Health Care Delivery Services
Gwen Dee Abeles
Erika Suzanne Berman

Florence L. Oppenheimer Surgery Prize
Laura Ann Karch

The Ellen Parker Memorial Award for
Excellence in Geriatrics
Tony Tsai

Dr. Howard R. Rappaport Pediatric Award
Usa-Gaye E. Robinson

Arthur Ross Award for Excellence
John Ladislav Bucek
Kathleen Jeanne Chen

Bela Schick Pediatric Society Prize
Akiko Kawamura

Nathan A. Selz Prize for Research in
Cardiovascular & Renal Diseases
Jianmin Chen
(Graduate School of Biological Sciences)
Waldo Feng
(Medical Scientist Training Program)

(Class of 1978)

Joseph R. Jagust Anesthesiology Award
Kenneth Brian Newman

George James Epidemiology Award
Bruce Goodkind Freeman

Nancy J. Goldin
Laurie Susan Teller Markin
Margaret Poole Mueller

Solomon Silver Award
Bruce D. Gelb, M.D.

Upjohn Clinical Excellence Award
Andrew D. Gotlin

Excellence-in-Teaching Awards

Patrick Eggena, M.D.
Art:hur Asher Kornbluth, M.D.
Edward J. Ronan, Ph.D.

Award winners are Class of 1993 except as noted

The

Mount Sinai Journal

of Medicine Awards 1992

Ralph Colp Prize

The 1992 Ralph Colp Prize for best surgical paper published in The Mount Sinai Journal
during the calendar year went to Leigh H. Nadler, MD, and Charles K. McSherry, MD, for their
article "Carcinoma of the Gallbladder: Review of the Literature and Report on 56 Cases at the
Beth Israel Medical Center" (vol. 59, no. 1 ). Dr. Nadler and Dr. McSherry are with the Depart-
ment of Surgery at the Beth Israel Medical Center. The award was established by colleagues and
friends of Ralph Colp, distinguished surgeon and for many years chief of surgery at The Mount
Sinai Hospital. All Journal awards are juried by a committee of five members of the publica-
tion's Editorial Board and are presented at the annual Prize Ceremony at the Mount Sinai
School of Medicine. The ceremony for the 1992 awards was held on May 18, 1993.

Globus Prize

The 1992 Globus Prize for the best paper with a clinical orientation published in this journal
during the calendar year went to the authors of two papers: Clive Rosendorff, MD, PhD, for
"Adrenergic Receptors in Hypertension and Heart Failure" (vol. 59, no. 4) and Robert A. Phillips,
MD, PhD, for "Pathophysiology and Treatment of Hypertensive Heart Disease" (vol. 59, no. 4).
Dr. Rosendorff is Professor and Associate Chairman, Department of Medicine, Mount Sinai
School of Medicine (CUNY), Chief, Medical Service, Veterans Affairs Medical Center, Bronx, NY,
and Director, The Mount Sinai Hypertension Program. Dr. Phillips is Director, Hypertension
Section, Associate Director, Cardiovascular Training Program, Division of Cardiology, Mount
Sinai School of Medicine (CUNY). This award honors Joseph H. Globus, founder and first
editor-in-chief (1934—1952) of The Mount Sinai Journal.

Student Prize

No award was made for the 1 992 Student Prize for the best student paper submitted to this
Journal during the calendar year.

Memoriam

Margit Freund Klemperer

Gabriel C. Godman, M.D.

We commemorate a great lady. For the first time
since the inception of this society of alumni and
alumnae of the Department of Pathology at
Mount Sinai Medical Center, Margit Freund
Klemperer is not with us. Many members will
recall having seen her, but those among us who
came to know her, even a little, will remember
her as a woman of exceptional presence, dyna-
mism and warmth. To the few of us who knew her
longest, Mitzi was a wise, compassionate, and
precious friend whose strength of character, hu-
manitarian impulse, and enormous integrity time
had enabled us to appreciate best. The story of her
life is inspiring; her wonderful relationship with
her husband Paul and her family and friends
were conjoined with significant professional
achievement and public-spirited action for good
causes, and we should here celebrate her success,
as we honor her memory.

Margit Freund Klemperer was born in
Budapest 93 years ago, the youngest of 13 chil-
dren of the chief rabbi of that city. Of the close-
ness, affection, and nurturing warmth of her fam-
ily life in Budapest she spoke to us often and with
feeling; her memories were so richly embellished
with anecdotal incident and detail that one was
transported in time and place to a far and happy
microcosm, forever lost. She spoke especially of
her father, whom she adored: indeed, an emanci-
pated and liberal man, yet observant (as was she),
he must have deserved her adulation. For, when
Mitzi and her sister decided to study medicine,
and in another country to boot . . . something un-
usual and unconventional for young women of
that day ... it was with his approval and support.
And so to Zurich they went, and soon after earn-

From a eulogy delivered to the alumni and alumnae of the
Department of Pathology on June 18, 1992. Address corre-
spondence to the author at 630 West 168th Street, P&S 11-
451, New York, NY 10032.

ing the degree, the sisters came to New York to
look for work. Mitzi applied at what was then the
New York Postgraduate Hospital to the young pa-
thologist, one Paul Klemperer; she got the job at
once, and a husband not many months later.

Theirs was an idyllic marriage; a lifelong
love affair, perpetuated by a mutual complemen-
tarity, support, and sharing which must be rare.
Mitzi, with her earthy, practical good sense, was
the perfect companion to her gentle, scholarly,
somewhat unworldly husband, wholly devoted to
science, medicine, and scholarship. Undemand-
ing, he was also woefully underpaid, even by the
standards of that day, and Mitzi's professional in-
come not less than her managerial skill in run-
ning the household were much needed. Her great-
est domestic accomplishment was to have raised
and educated two remarkable sons, both of them
illustrious: Bill, the eldest, is Hersey Professor of
Chemistry at Harvard, a distinguished physics
chemist, and a member of the National Academy
of Sciences; Martin, who is temperamentally
more like Paul, is the much admired Professor of
Pediatrics now in Tampa. Whether by genes or
nurture, they could hardly be other than men of
great character.

Mitzi had to commute all the way to St. Vin-
cent's Hospital in Staten Island, where she served
as pathologist, bacteriologist, and clinical chem-
ist. It says something about her that the nuns
who ran the hospital, at first suspicious and aloof,
became her cordial, lifelong friends, and it was
the sister superior who motivated the trip of Paul
and Mitzi to Ireland in the 1950s. That trip was
one of many journeys that Paul and Mitzi made,
always together ... to Europe, Russia, Israel,
Asia, South America . . . alternating between
where he had to go and what she wanted to ex-
plore. She continued to travel, alone and indomi-
table, especially to Israel, until the age of 91. But
the place that she and Paul loved most of all was
the spacious old homestead and farm in Nelson,

340

The Mount Sinai Journal of Medicine Vol. 60 No. 4 September 1993

Vol. 60 No. 4

MARGIT FREUND KLEMPERER— GODMAN

341

NH that they acquired in the 1940s for a pittance
as a ramshackle relic and lovingly built up over
the years. They became accepted, integral, be-
loved members of that old closeknit New England
community of flinty Yankees and fixtures of Nel-
son Village Old Home Day. When Paul died in the
mid 1960s, a memorial service was held in the
Nelson village church in midwinter, and they
brought someone to recite the memorial prayer
. . . the Kaddish ... in Hebrew. The ladies from
the village, her neighbors the Tuttles, the Ly-
mans, the Joneses, the Tolmans who loved Mitzi,
would visit on summer afternoons bringing gifts
from gardens, and be visited in turn, with prod-
ucts of her orchard or bake oven.

From an honorable career in pathology, Mitzi
switched (rather radically) to the practice of child
psychiatry (in the 1950s), again with signal suc-
cess, and when she relinquished that practice she
returned to pathology, reading biopsies here at
Mount Sinai until the age of 83, capable, smart,
and respected, an excellent diagnostician, and ex-
pert in otorhinolaryngeal and dental pathology.

She was passionately engaged in public and
liberal causes, and in the 1950s and 1960s letters
signed Margit Freund Klemperer were showered
on newspapers, and printed. Remembered with
pride, not only by the beneficiaries, are — most of
all — the efforts of Paul and Mitzi Klemperer on
behalf of emigres fleeing from Nazi Europe in the
1940s, their staunch protection of Dr. Otani and
his family during the shameful anti-Japanese fe-
ver here in 1942, their activism against the Mc-
Carthy period outrages and on behalf of its vic-
tims, their concern for Israel.

Independent, self-reliant, capable to the end,
she lived alone in that lovely apartment on 5th
Avenue overlooking the park that had been the
site of so many warm-hearted, convivial, and ed-
ucational gatherings, so many cordial conversa-
tions. And she kept it just as it was when Paul
was alive, always fresh and flowered. It was there
that she died, suddenly and unexpectedly, on
April 8, 1992, as she would have wished; she is
mourned, and celebrated as a still beneficent
presence, by those fortunate to have known her.

The Department of Rehabilitation Medicine
The Mount Sinai School of Medicine (CUNY)

The Page and William Black
Post-Graduate School of Medicine

The Rehabilitation Medicine Services
Veterans Affairs Medical Center, The Bronx, NY

sponsor the first international course on

myofascial
PAIN

diagnosis & treatment

emphasizing practical clinical management

course: December 8-11, 1993
workshop: December 12, 1993
The Mount Sinai Medical Center
100th St. & Madison Ave., New York, NY
Stern Auditorium

course director: Andrew A. Fischer, MD, PhD, Chief, Rehabilitation Medicine Service, VAMC, Bronx, NY
faculty: Janet Travell, MD, Hans Kraus, MD, and other leading authorities

Fees:

course: $390 physicians

$290 others
workshop: $120 physicians
$100 others

CME credits available

designed for pain specialists, including physiatrists,
orthopedists, anesthesiologists, rheumatologists,
neurologists, physical and occupational therapists,
chiropractors, and myotherapists

discounted hotel and travel arrangements for registrants: Zenith Travel (212) 241-9447

for information call The Page and William Black Post-Graduate School of Medicine, Mount Sinai
School of Medicine, Box 1193, One Gustave L. Levy Place, New York, NY 10029

(212) 241-6737

Because One Size Doesn^t Fit All...

Eli Lilly and Company can suit all your
needs with the most complete line of human
insulins available.

Featuring HumuHn 70/30* and our latest
addition to the premized line, HumuHn 50/50t
-especially useful in situations in which
a greater insuhn response is desirable for
greater glycemic control.

Hiunuliri ®

human insulin
[recombinant DNA origin ]

Tailor-made options in
insulin therapy

WARNING: Any change of insulin should be made cautiously
and only under medical supervision.

* Humulin® 70/30 (70% human insulin isophane suspension,
30% human insulin injection |recombinant DNA origin]).

tHumulin* 50/50 (50% human insulin isophane suspension,
50% human insulin injection [recombinant DNA origin]).

Global Excellence in Diabetes Care
Eli Lilly and Company

Indianapolis, Indiana
46285

Required Forms for Authors:

AUTHORSHIP RESPONSIBILITY, FINANCIAL DISCLOSURE, AND ASSIGNMENT OF COPYRIGHT

Each author must read and sign (1) the statement on authorship responsibility: (2) the statement on financial disclosure; and (3) either the statement
on copyright transfer or the statement on federal employment. If necessary, photocopy this document to distribute to coauthors for their signatures.
Please return all copies to the address below.

1. Authorship Responsibility

I certify that I have participated sufficiently in the conception and
design of this work and the analysis of the data (where applicable), as
well as the writing of the manuscript, to take public responsibility for it.
I believe the manuscript represents valid work. I have reviewed the final
version of the manuscript and approve it for publication.

Author(8) Signature(s)

Neither this manuscript nor one with substantially similar content under
my (our) authorship has been published or is being considered for
publication elsewhere, except as described in an attachment.

Furthermore, I attest that I shall produce the data on which the manu-
script is based for examination by the editors or their assignees should
they request it.

Date Signed

2. Financial Disclosure

I certify that I have no affiliation with or financial involvement in any
organization or entity with a direct financial interest in the subject
matter or materials discussed in the manuscript (eg, employment, con-

Author(8) Signature(8)

sultancies, stock ownership, honoraria), except as disclosed in an at-
tachment.

Any financial project support of this research is identified in a note on
the title page of the manuscript.

Date Signed

3. Copyright

In compliance with the Copyright Revision Act of 1976, effective Jan-
uary 1, 1978, The Mount Sinai Journal of Medicine, in consideration of
taking further action in reviewing and editing your submission, requests
that each author sign a copy of this form before manuscript review can
proceed. Such signature shall evidence the mutual understanding be-
tween The Mount Sinai Journal of Medicine and the undersigned au-

Author(8) Signature(8)

thor(s). thereby transferring, assigning, or otherwise conveying all copy-
right ownership, including any and all rights incidental thereto, exclu-
sively to The Mount Sinai Journal of Medicine.

In consideration of the Journal's action in reviewing and editing this
submission, the author(s) undersigned hereby transfer(s) or otherwise
convey(s) all copyright ownership to the Journal if such work is pub-
lished by the Journal.

Date Signed

US Federal Employees: 1 was an employee of the US federal govern-
ment when this work was conducted and prepared for publication; there-

fore, it is not protected by the Copyright Act and there is no copyright;
thus, ownership cannot be transferred.

Author(8) Signature(8)

Date Signed

Return the original signed form to Sherman Kupfer, M.D., Editor, The Mount Sinai Journal of Medicine, Box 1094, 50 E. 98th Street,
New York, NY 10029. Retain one copy for your flies. You may make as many photocopies as you need if your paper has more than
three authors.

Instructions for Authors

THE MOUNT SINAI JOURNAL OF MEDICINE, established in 1934 as The Journal of the Mount Sinai Hospital, is a peer-re-
viewed general medical journal. It is indexed or abstracted in Index Medicus, Current Contents, Science Citation Index, Hospital
Literature Index. International Nursing Index, Excerpta Medica, Chemical Abstracts, Biological Abstracts, and major databases.
Theoretical and basic science articles for the clinician, reviews, reports of clinical or research experience, articles on any subject
affecting the practice of medicine, short notes in any specialty, book reviews, letters, and articles by medical students are
welcome. The Journal is a participant in the Agreement on Uniform Requirements for Manuscripts Submitted to Biomedical
Journals. The Journal aims at responding to authors of manuscripts within 8 weeks of date of submission.

Authorship Responsibility, Financial Disclosure, and Assign-
ment of Copyright. Each author is required to sign the three
statements on the accompanying page.

Manuscript Preparation. Submit manuscripts in triplicate,
including three sets of illustrations. Manuscripts must be
typewritten, double-spaced throughout (including refer-
ences, legends, and tables), on one side of 8^/2 x 11-in. white
bond paper, with generous margins. Number pages consecu-
tively. Begin a new page for title page, abstract, first page of
text, references, each table, and legends. Computer-generated
typescripts must be double-spaced and must be easy for ed-
itors and typesetters to read. Computer tapes or disks: consult
managing editor.

Title Page. On the title page, on separate lines (double-
spaced), include (1) title of article, phrased as concisely as pos-
sible; (2) name of each author, including first name and de-
gree; (3) name and address of departments and institutions
from which the work originated; (41 acknowledgments of
sources of support; (5) name, place, and date of any confer-
ences at which paper has been delivered; (6) the statement
Address reprint requests to ... , giving full name and ad-
dress, with zip code, of the appropriate author.

Key Words. Supply 3 to 10 key words or phrases at the bottom
of the title page, for use in indexing the article. Use Med-
ical Subject Headings from Index Medicus.

Units of Measurement. Use SI units, giving range or interval
of normal values.

Illustrations. Photographs, line drawings, graphs, and charts
should increase understanding of the text. Line drawings,
graphs, and charts should be professionally prepared. All il-

lustrations should be submitted in triplicate as sharp, high-
contrast glossy prints. Photomicrographs must be 5'/h in.
wide. On the back of each print affix a gum label with the
number of the figure (numbered consecutively in arable nu-
merals as cited in text), title of manuscript, name of senior
author, and the word "top" with an arrow indicating top edge.
Illustrations which are to appear together should be mounted
accordingly as plates and should correspond to each other in
size. Protect the prints by enclosing them in heavy cardboard;
do not use paper clips. Photographs of patients can be pub-
lished only if a copy of the permission is submitted with the
photograph. Eliminate names of patients and hospital
numbers from x-rays.

1. Nadelman RB, Wormser GP. A clinical approach to Lyme
disease. Mt Sinai J Med 1990;57:144-156.

2. International Committee of Medical Journal Editors. Uni-
form requirements for manuscripts submitted to biomedical
journals. N Engl J Med 1991;324:424-428.

When a manuscript is accepted for publication, you will be
asked to confirm the accuracy of all references.

Reprint Orders. A reprint order form is sent to the corre-
sponding author with proofs; return it to the editorial office.

THE

MOONT SINAI JOaRNAL
OF MEDICINE

Forthcoming:
Grand Rounds.

Gene Therapy for Immunodeficiency and Cancer

R. Michael Blaese
National Institutes of Health

Neonatal Herpes Simplex Virus Infections:
Natural History, Pathogenesis, and Preventability

Richard J. Whitley

University of Alabama School of Medicine

Controversies in Screening for Prostate Cancer

Martin Resnick

Case Western Reserve University School of Medicine

The Function of the p53 Tumor Suppressor
Gene and Its Clinical Correlates

Arnold Jay Levine
Princeton University

Mechanisms of Autoantibody Production and
Their Role in Disease

Keith Elkon

The Hospital for Special Surgery

Pathway for Water Absorption in
Isosmotic Transporting Epithelia

Guillermo Whittembury

Institute Venezolano de Investigaciones Cientificas

Theme Issues

OCTOBER 1993

GUiality Improvement

Editor: Suzanne Kramer

NOVEMBER 1993

Update on Treatment of the Elderly
On the 10th Anniversaiy of the
Department of Geriatrics

Editors: Myron Miller and Rein Tideiksaar

JANUARY 1994

Transplantation

Editor: Lewis Burrows

MARCH 1994

Cervical Disk Disease:
Controversies in Neurosurgeiy

Editors: Kalmon Post and Martin H. Savitz

Available now:

Festschrift for Rosalyn 8. Yalow: Issues in

Hormones, Metabolism, and Society Medical Ethics

Eugene W. Straus, editor Daniel A. Moros and Rosamond Rhodes, editors

volume 59. no. 2. March 1992 volume 60. no. 1, January 1993

96 pages $15 84 pages *15

Subscriptions for 1993: $65 individuals, US.: $70 all libraries: $75 individuals outside US.

To order subscriptions or copies of back issues ($15 each), please send check, payable to The Mount Sinai Journal of Medicine, to the Journal at Box 1094, One Gustaoe L Levy Plac
New York, NY 10029-6574: counter sales at Suite IE, 50 E. 98th Street. New York, NY (phone: 212 241-6108: FAX: 212 722-6386).

Toward the Conquest of Pain

Allan P. Reed, editor
volume 58. no. 3. May 1991
84 pages $15

OGNT SINAI JOGRNAL

F MEDIQNE

)LaME 60 NUMBER 5

OCTOBER 1993

-y —I

5:r

Perspectives on Quality
Improvement in Health Care

Suzanne Carter Kramer, guest editor

From The Mount Sinai Hospital, New York

BtlH

/VCV Veterans
Vfc^ Affairs

Israel

Medical

Center

T^T TIT The Mount Sinai Journal of Medicine is published by The Mount Sinai Medical Center of
I II New York and has the following affiliates: Beth Israel Medical Center, New York; Bronx

MOUNT SINAI
JOURNAL OF
MEDICINE

Veterans Affairs Medical Center, New York; and Elmhurst
Hospital Center, New York.

Editor-in-Chief

Sherman Kupfer, M.D.

Editor Emeritus

Lester R. Tuchman, M.D.

Associate Editors

Harriet S. Gilbert, M.D. Julius Wolf, M.D.

Managing Editor

Claire Sotnick

Business and Production Assistant

Karen Schwartz

Assistant Editors

Stephen G. Baum, M.D.
David H. Bechhofer, Ph.D
Constanin A. Bona, M.D.,
Edward J. Bottone, Ph.D.
Jurgen Brosius, Ph.D.
Lewis Burrows, M.D.
Joseph S. Eisenman, Ph.D.
Adrienne M. Fleckman, M.D
Richard A. Frieden, M.D.
Steven Fruchtman, M.D.
Paul L. Gilbert, M.D.
James H. Godbold, Ph.D.

Richard S. Haber. M.D.
Noam Harpaz, M.D.
Ph.D. Dennis P. Healy, Ph.D.
Tomas Heimann. M.D.
Barry W. Jaffm. M.D.
Andrew S. Kaplan, D.D.S.
Samuel Kenan, M.D.
Suzanne Carter Kramer. M.Sc.
Mark G. Lebwohl, M.D.
Kenneth Lieberman, M.D.
Charles Lockwood, M.D.

Lynda R. Mandell, M.D., Ph.D.

Steven Markowitz, M.D.

Bernard Mehl, D.P.S.

Myron Miller, M.D.

Edward Raab, M.D.

Allan Reed, M.D.

Allan E. Rubenstein, M.D.

David B. Sachar, M.D.

Henry Sacks, M.D.

Robert Safirstein, M.D.

Ira Sanders, M.D.

Editorial Board

Barry Freedman, M.B.A.
Richard Gorlin, M.D.
Nathan Kase, M.D.

Panayotis G. Katsoyannis, Ph.D.
Charles K. McSherry, M.D.
Jack G. Rabinowitz, M.D.

John W. Rowe, M.D.
Alan L. Schiller, M.D.
Alan L. Silver, M.D.

Alvin S. Teirstein, M.D.
Rosalyn S. Yalow, Ph.D.

The Mount Sinai Journal of Medicine (ISSN No. 0027-2507; USPS 284-860) is published 6 times a year in January, March, May.
September, October, and November in one indexed volume by the Committee on Medical Education and Publications (Owner), The
Mount Sinai Hospital, Box 1094, One Gustave L. Levy Place, New York, NY 10029-6574. Subscription price: individuals, U.S.,
$65 per year; libraries, $70; individuals outside the U.S., $75. Single copies, $15. New subscriptions begin with the first issue of
the calendar year. Send notice of change of address 60 days before the change is effective. Second-class postage paid at New York,
NY and at additional mailing offices. POSTMASTER: Send address changes to The Mount Sinai Journal of Medicine, Box 1094, One
Gustave L. Levy Place, New York, NY 10029-6574. Copyright 1993 The Mount Sinai Hospital.

THE

MOGNT SINAI JOGRNAL
OF MEDICINE

Volume 60
Number 5
October 1993

CONTENTS

PERSPECTIVES ON QUALITY IMPROVEMENT IN HEALTH CARE

Suzanne Carter Kramer, editor

Introduction Suzanne Carter Kramer 346

A Brief Thesaurus of Quality Robert Matz 348

Quality, Deming's Principles, and Physicians Louis Evan Teich-

holz 350

Risk Management in Health Care: Where Did It Come From and
Where Is It Going? Barbara Challan, Jane McConnell, and
Alice Walsh 359

A Century of Progress: Nursing Quality Improvement at The Mount
Sinai Medical Center Linbania Jacobson, Mary O'Connell,
Carol Nicol O'Brien, and Gail Kuhn Weissman 363

Social Work Accountability: A Key to High-Quality Patient Care
and Services Helen Rehr, Susan Blumenfield, Alma T.
Young, and Gary Rosenberg 368

Patient Representation as a Quality Improvement Tool Ruth

Ravich and Lucy Schmolka 374

Prevention of Medication Errors: Developing a Continuous-Qual-
ity-lmprovement Approach Keith Bradbury, Julie Wang,
George Haskins, and Bernard Mehl 379

Quality Improvement in Action: A Falls Prevention and Manage-
ment Program Sylvia M. Barker, Carol Nicol O'Brien, Dor-
othy Carey, and Gail Kuhn Weissman 387

Research in the Hospital Setting on Human Subjects: Protecting

the Patient and the Institution Ruth Scheuer 391

The Consumer Survey Review Process: A Pathway to Quality

Edward J. Speedling, Arthur A. Nizza, Suzanne Eichhorn,

Gary Rosenberg, and Phyllis Schnepf 399

Quality Improvement: A Review of Five Books and a Personal

Overview Charlotte Muller 405

GRAND ROUNDS

Controversies on Screening for Cancer of the Prostate Martin

Resnick 412

ABSTRACTS

Research Day in Medicine 1993 417

Index to Advertisers

Course on Myofascial Pain
Diagnosis and Treatment

Page 451

Post-Graduate School of Medicine
Continuing Medical Education Courses

Page 453

\
1
1

■

A/hen you doh'l know all

he medicines your patients are taking,

)rotecting their health

s a crap shoot. y #^

1 -::

Counseling patients about all their
medicines improves their odds of
getting well and slaying well.

EVERYONE WINS WHEN YOU TALI
^^end me a free Medicine Counseling Kit '

OrganizatiL

C.i> Sldic Zip

Mail lo:

National Council on Patient Information

and Education
666 Eleventh Street. NW, Suite 810
Washington. DC 20001
To fax your request - (202) 638-0773

this publicQtion
is QVQiloble in
micfotofm

Rii -■■■■BI--5I

Please send me additional information.

University Microfilms International

300 North Zeeb Road 18 Bedford Row

Dept. P.R. Dept. P.R.

Ann Arbor, Ml 48106 London, WCI R 4EJ

U.S.A. England

Name

Institution

Street

City

State

-Zip

Perspectives On
Quality Improvement In
Health Care

Introduction

Suzanne Carter Kramer,
M.Sc.

Associate Director

The Mount Sinai Hospital

Perspectives
On

Quality
Improvement
In Health Care

This theme issue of The Mount Sinai Journal of Medicine focuses
on quality improvement in health care, a discipline that has un-
dergone significant change during the last several decades. As the
nation enters a new era of health care reform, the issues associated
with quality in health care will become more important then ever
before. Health care providers will increasingly be focusing their
efforts on issues of quality juxtaposed with issues of fiscal con-
straint.

Quality assurance in health care has evolved dramatically,
from the traditional outcome-oriented quality monitoring ap-
proach to newer process-oriented approaches of total quality man-
agement and continuous quality improvement. This pervasive
shift in philosophy is most evident in the new standards of the
Joint Commission on the Accreditation of Healthcare Organiza-
tions (JCAHO), which require all health care institutions to incor-
porate elements of continuous quality improvement in their qual-
ity assurance programs, while not abandoning the more
traditional quality assurance and monitoring tools. The articles
included in this issue provide an overview of the evolution of qual-
ity improvement in health care, and in addition describe diverse
quality improvement approaches adopted by many disciplines in
the academic health sciences center setting. R. Matz's article pro-
vides a useful introduction in its discussion of quality improve-
ment methods and practices.

In his article "Quality, Deming's Principles, and Physicians,"
L. Teichholz explores the roots of the current revolution in health
care quality in concepts developed for manufacturing and other
industries and their subsequent application to health care organi-
zations. Additionally, he outlines the development and implemen-
tation of continuous quality improvement in health care organiza-
tions and its foundation in the tools developed for industry by
Deming, Juran, and others. The final section of this article ad-
dresses the critical importance of investing physicians in a suc-
cessful quality improvement process. B. Challan et al.'s article on
health care risk management explores one important component of
the overall quality assurance function and its current role and
scope in health care organizations. As the article indicates, the role
of the risk management team will only grow in importance as
health care technology continues to evolve, further expanding the
ethical issues and appropriate boundaries of medical science.

An additional set of articles, by Jacobson et al., Rehr et al.,
Ravich et al., and Bradbury et al., describes the historical devel-
opment and key elements of the quality assurance function in a
number of disciplines and clinical areas, including nursing, social
work, patient representation, and pharmacy. Each article dis-
cusses innovations in quality monitoring, evaluation, and im-
provement that have been developed at The Mount Sinai Medical
Center and, in many cases, have set the example for their respec-
tive fields. Weissman et al.'s article on falls prevention provides a
specific example of "quality improvement in action," describing the
identification of a major, often preventable, patient care problem
and the process and systems put in place to reduce its incidence.
The article by R. Scheuer on conducting research on human sub-
jects also addresses a specific institutional function in the context
of quality assurance, highlighting the importance of applying risk
management and quality assurance principles to the functioning of

346

The Mount Sinai Journal of Medicine Vol. 60 No. 5 October 1993

Vol. 60 No. 5

INTRODUCTION— CARTER KRAMER

347

a hospital's institutional review board, the body
charged with protecting the rights of research
subjects. Finally, the article by Speedling et al.
examines the development and application of one
specific methodology for measuring and improv-
ing upon the quality of patient care in a hospital
setting.

The issue closes with a series of book reviews
by C. Muller on a variety of topics related to qual-
ity of care. The books reviewed cover a broad
range of topics, including development of clinical
practice guidelines, implementation of continu-
ous quality improvement programs, development
of nursing quality assessment programs based on
the new JC AHO standards, and needs assessment
and outcome measures for primary care.

In addition to the mounting body of literature
advocating continuous quality improvement and
total quality management, a number of other
health care quality trends have emerged, includ-
ing the growing practice of benchmarking, a
growing emphasis on the development of clinical
practice guidelines, and increased analysis of
data on outcomes. All are being used as a means

of promoting both better-quality and more cost-
effective health care. Although the articles in this
issue do not explicitly address these trends, they
are gaining greater prominence in the changing
health care environment, particularly as man-
aged care models of health care delivery increas-
ingly dominate the marketplace.

This theme issue is intended to provide the
reader with an understanding of how quality as-
surance has evolved in health care in recent years
and the multiplicity of approaches adopted in an
academic health sciences center for evaluating
and achieving quality patient care. As private
sector, state, and federal initiatives continue to
emphasize fiscal constraint, health care prices
will eventually stabilize. Quality, however mea-
sured or defined, will then become the most sig-
nificant competitive factor in the provision of
health care. A commitment to quality in all areas
of the academic health sciences center is therefore
a strategic imperative for the future.

Special thanks to Jodi Katz, M.Sc, for her
editorial assistance with this publication.

A Brief Thesaurus of Quality

Robert Matz, M.D.

Quality assurance doesn't! By this I mean that
quality assurance as previously practiced in
health care never assured the quality of the care
provided.

The approach involved inspection of a final
"product" — a medical record or a bad outcome —
and made the assumption that deviations from
some preconceived outcome were the result of bad
providers. In this mode, the quality assurance
process attempted to ferret out the bad apples —
the "gotcha" approach — and thereby improve
overall quality of care. Although this process oc-
casionally resulted in the removal of a few poor
practitioners, the ultimate effect on quality of
care was minimal. Further, the process engen-
dered a defensive, hostile response from compe-
tent health care professionals who were doing
their best day in and day out. Much of what was
done in the name of quality was paper compliance
with outside regulations. It supported the precept
that there is nothing so useless as doing some-
thing perfectly that doesn't need to be done.

The mechanism of traditional quality assur-
ance was primarily via the isolated bad occur-
rence, the one-time audit, or the anecdotal inci-
dent. Although occasionally necessary and still
appropriate, the traditional approach, when used
as the sole modality, creates a confrontational at-
mosphere inimical to real improvement. It espe-
cially runs counter to the scientific training and
method that has shaped American medicine for
the past 70-100 years.

Modern medical care is based on the concept
of continuous quality improvement; to the practi-

From the Department of Medicine, Mount Sinai School of
Medicine, and the Department of Quality Assurance and Uti-
lization Review, The Mount Sinai Hospital, New York, NY.
Address reprint requests to the author. Medical Director of
Quality Assurance and Risk Management, Box 1233, The
Mount Sinai Medical Center, One Gustave L. Levy Place, New
York, NY 10029.

348

tioner there is nothing new about it except the
name. It avoids reliance on isolated case reports,
which, however interesting and educational, may
not reflect the bigger picture. In the face of a prob-
lem, a hypothesis is generated, data is collected
and analyzed, control trials are undertaken, re-
sults are subjected to statistical analyses, discus-
sion is undertaken, conclusions are reached, and
recommendations for or against certain actions
are made. Once these recommendations are im-
plemented, the results are monitored, resulting in
the acceptance or rejection or fine-tuning of the
hypothesis, as appropriate. This overly simplified
analysis of how we have progressed since the
Flexner report is familiar to everyone involved in
medical care and is now, finally, the approach be-
ing applied to the analysis of quality in health care.
Unfortunately, the path has been a difficult, frus-
trating, and painful one, and we have had to re-
leam our own methodology from industrial models.

A basic tenet of this new paradigm of quality
improvement, now being called "total quality
management" or "continuous quality improve-
ment," is the need for "doing it with data" — "In
God we trust; all others must show us their data."
In other words, anecdotes, pontifical pronounce-
ments, and guesses can no longer be the basis for
continued practice. In order to understand and
achieve quality, the problem must be defined in
quantitative, measurable terms.

Another concept which has evolved is "doing
the right thing right the first time, every time."
This encapsulates the entire purpose of continu-
ous quality improvement, and every word of the
statement is critical. It means not waiting until
the end of the process to "inspect" the product and
then reject it if it is not right, but rather building
quality into the process and product from the very
beginning. Quality assurance by inspection is
wasteful and costly and dangerous.

The new paradigm of continuous quality im-
provement assumes that everyone comes to work

The Mount Sinai Journal of Medicine Vol. 60 No. 5 October 1993

A THESAURUS OF QUALITY— MATZ

349

every day to do the best job of which they are
capable, given their education, available re-
sources, the current system of doing things, and
the material they have to work with. With this in
mind, the approach to improving the outcome is
to look at the system or process and attempt to
continuously improve its elements, thereby en-
abling everyone to do the right thing in the right
way the first time, every time.

Another valuable precept is that processes or
systems are never perfect. They may statistically
approach perfection, but all processes fluctuate
around some statistical mean, and tinkering with
a process which is "in control" or working may be
inappropriate. In order to look at these systems or
processes in a meaningful way, however, we must
understand the process and collect sufficient data
to enable us to analyze it with a reasonable de-
gree of statistical validity. There will always be
some variation around the mean, and this may or
may not warrant our attention. The point is that
establishing a threshold for a process and then
not looking at it if the threshold is not violated is
no longer acceptable. If we meet or even exceed
our "threshold," the next step in continuous qual-
ity improvement is the "benchmark" — a point of
reference which serves as a standard by which the
process may be measured. This is conceptually
the high water mark — the best in the business —
for which we all strive. And if we get there, we
should try to exceed it. That is quality improve-
ment.

But beware; "Perfection is a city in Russia
and nobody wants to live there." There may come
a point in the improvement process when the next
step is just too costly, too difficult, or too unim-
portant to warrant taking it. It is the journey to-
ward perfection which is important for most pro-
cesses, the step-by-step continuous process of
measuring, analyzing, and monitoring, and —
when necessary — changing to reach and exceed
the benchmark. As David Kearns, chief executive
officer of Xerox, has so aptly put it, "In the race
for quality there is no finish line."

How should we define "quality"? Quality is
what our customers expect of us, and the only way
to know what they expect is to ask them. Why do
I say "customers" instead of patients? Because
whether we are physicians, nurses, social work-
ers, or technicians, we all have many individuals

(not just patients) who expect something of us in
this complex process called health care. The phy-
sicians' orders must be clear and timely in order
for the nurse to pick them up and request trans-
port to take the patient to x-ray. At every step in
this process we each are attempting to satisfy —
yes — "customers." The only way to know whether
we are meeting their expectations is to ask; pa-
tient satisfaction surveys and interdisciplinary
quality improvement groups begin this struc-
tured analysis. Only by this multidisciplinary
process can we define what quality is. Quality is
never accidental. It is the result of thoughtful de-
sign. One definition which I like is, "Quality is
meeting or exceeding customers' expectations
100% of the time."

How can we evaluate such a complex system,
find its flaws, and modify and improve it? By ask-
ing the people who actually perform these activ-
ities what they do, how they see the process, how
they would measure it, and what they think could
be done to improve it. Then we must permit them,
within limits, to act. This requires a willingness
on the part of management to listen to employ-
ees— a process that can be threatening to middle
management, though it need not be. Most impor-
tant, it requires a commitment from the top of the
organization such that the process of quality im-
provement will become a part of the ethic of the
institution, that employees will be empowered to
seek processes that need improvement and, where
appropriate, to implement the recommended ac-
tions. It means an organizational commitment to
quality improvement as job one.

This is a top-to-bottom approach which takes
time and resources — no quick fixes, no one-time
audits, but a continuous, open-ended process of
looking and improving. In most instances where
continuous quality improvement has been incor-
porated into the institutional structure it has, in
the long run, been cost effective.

The papers in this symposium reflect the
spectrum of the changing face of how quality is
looked at, measured, and ultimately improved in
the hospital setting. They cover many disciplines
in the health care process. This is only a begin-
ning. As evidenced by the high quality and broad
range of contributions, the future looks promising
for the application of these new tools in the quest
for continual improvement.

Quality, Deming's Principles,
and Physicians

Louis Evan Teichholz, M.D.

Quality

In the 1980s, driven by standards from the
Joint Commission on Accreditation of Healthcare
Organizations (JCAHO) and by external regula-
tory agencies in the various states, the concept of
quality assurance became an important concern
for hospitals in the search for quality. The quality
assurance approach in the health care field has
concentrated on analysis of structure, process,
and outcome. Much of the emphasis has been on
outcome — what happens to patients. This concept
of quality assurance is based on examination of
poor outcomes and to a large extent on the "bad
apple" theory, which implies that the problem lies
in the individual and that people must be made to
care. The emphasis has been on quality problems
as faults of people rather than of the process itself
(1). This quality assurance process has been de-
scribed as static (2), adversarial, and punitive,
thus fostering a "we" versus "they" attitude. It
has been reactive rather than proactive and based
on the concept that quality can be improved by
inspection and overseeing rather than by chang-
ing the process (3). Separate departmental qual-
ity assurance staffs have been developed, which
can undermine institutional teamwork (4), and in
many hospitals the quality assurance department
has been perceived as a policeman. Most impor-
tant, most physicians have lacked interest and
scanted involvement in this important process be-
cause of the adversarial stance and because phy-
sicians are in many cases outside the process (5).

Despite much experience with this approach,

From the Division of Cardiology, Department of Medicine,
The Mount Sinai Medical Center, New York, NY. Address
reprint requests to the author at the Division of Cardiology,
Box 1030, Mount Sinai Medical Center, One Gustave L. Levy
Place, New York, NY 10029.

whether it has improved the quality of medical
care in any major way is not clear. Certainly no
evidence exists of increased patient satisfaction;
and no significant improvement appears to have
occurred in another dimension of improvement,
patients injured in hospitals: a California study
conducted in the 1970s found that 0.82% of 20,000
hospital patients showed some disability believed
to be caused by negligence (6), and a Harvard
study conducted in 1984 of 30,195 patients in
New York State found that 1% had meaningful
disability deemed to be caused by negligence (7).

Quality is the end result of a complex inter-
action of people and support systems (5). The cur-
rent process has not gone much beyond superfi-
cial problem analysis and finger-pointing.
However, in the 1990s the focus has started to
change. Although outside regulatory agencies are
still applying pressure on quality assurance and
reporting of poor outcomes and complications, we
are entering a new era in which, clearly, the way
to improve quality is to improve the process itself
The role of the quality assurance department
must change from being a cop to being a "coach"
(4), and quality must be totally integrated within
the hospital management system.

The new revolution is based on the pioneer-
ing work on quality in industry by Deming, Ju-
ran, and others, initially introduced in Japan and
now being applied to American industry (8-15).
Berwick (3, 16) and others (2, 17-20) have applied
these industrial concepts to the health care set-
ting, and the term "continuous quality improve-
ment" (CQI) has emerged. Another term usually
used synonymously with CQI is "total quality
management" (TQM). The focus in CQI is the pro-
spective analysis of statistical data and trends,
not — as with the older concept — retrospective re-
view of individual cases with poor outcomes. The
emphasis now is on finding common denomina-

350

The Mount Sinai Journal of Medicine Vol. 60 No. 5 October 1993

Vol. 60 No. 5

QUALITY, DEMING'S PRINCIPLES, AND PHYSICIANS— TEICHHOLZ

351

tors and asking what the overall process is and
whether it can be improved, not on what went
wrong in any individual case.

The Hospital Corporation of America has ap-
plied Doming techniques to hospital organization
(10) and various hospitals — the University of
Michigan hospital (21) is one — have described
their experiences with this technique (20-26).
The applications of CQI to improving physician
care have recently been reviewed (27).

Most importantly, the National Demonstra-
tion Project on Quality Improvement and Health
Care was developed in September 1987 to answer
the question, "Can the tools of modern quality
improvement, with which other industries have
achieved breakthroughs in performance, help in
health care as well?" (28). This National Demon-
stration Project "achieved the goal of showing
that the concepts and tools in industrial quality
improvement can be used to improve the quality
of processes in the health care industry. . . . The
familiar tools of industrial quality improve-
ment . . . [have been] successfully applied to a
wide variety of health care processes" (28).

The process of continuous quality improve-
ment encompasses a multiplicity of concepts, in-
cluding statistical quality control and a new man-
agement philosophy (2, 3, 18, 20, 25, 28). The
JCAHO has adopted CQI as an important new
concept and is revising its standards and review
to reflect the new principles (29). A major goal is
to modify most standards to reflect the basic con-
cepts of CQI by 1994 (30). In fact, a recent poll of
hospital chief executive officers indicated that al-
most 60% are now implementing a CQI/TQM pro-
gram and, of those without a program, 75% were
planning to start one within the next year (31).
The American Hospital Association has recently
published three books to help hospitals and phy-
sicians turn theory into practice (32-34).

Deming's Principles

W. Edwards Deming was born on October 14,
1900 and grew up on a Cody, Wyoming home-
stead. He received an undergraduate degree from
the University of Wyoming in 1917, a master's
degree in mathematics from the University of
Colorado, and a Ph.D. in physics from Yale Uni-
versity. He initially worked at the Hawthorne
plant of Western Electric, and joined the U.S. De-
partment of Agriculture in 1927. During subse-
quent years, he was introduced to Dr. Walter
Shewhart at Bell Laboratories and became in-
volved in learning and understanding statistical
control methods. He also studied statistics with

FLOW CHART

CONTROL CHART

CAUSE AND EFFECT

PARETO CHART

Fig. I. Four important graphic tools for statistical control
processes: flow chart showing each step in a process; cause-
and-effect or Ishikhawa diagram; control chart to analyze
variation over time of any parameter; Pareto bar chart deliri-
eating frequency of each factor contributing to an effect. X,
mean; UCL, upper control limit; LCL, lower control limit; ar-
row, data point outside control limits.

Dr. Ronald Fisher. Because of his expertise in sta-
tistical control and sampling methods, he was en-
listed to develop statistical methods for the 1940
census and, during the war years, he taught sta-
tistical methods and quality control. In 1947 he
was again enlisted, this time by the Supreme
Command Allied Powers, to prepare for the 1951
Japanese census, and thus began his long associ-
ation with Japanese industry.

Deming taught the Japanese his methods of
statistical control and quality improvement and
became widely known in Japan, revolutionizing
their quality process in industry. However, his
methods and philosophy were not well known in
the United States, nor was there any widespread
application in American industry until after a
1980 NBC television special about his techniques
in Japan. His book, Out of the Crisis (35), pub-
lished in 1986, served as the foundation for the
current revolution. Deming proposed and dis-
cussed applications of his methods to the health
care setting in that book. Public discussion of ap-
plication of his principles to health care began
about two years later, in 1988, and has gathered
steam since (20, 36-39).

Statistical Methods. The major contribution
of Deming is the concept of statistical control pro-
cesses and the application of statistical theory
and methods to the improvement of quality. In
addition, in Out of the Crisis, he proposed 14
points which have broad implications for industry
in general and health care in particular, both for
overall organizational structure and goals and for
quality improvement itself.

352

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

Statistical quality control is a tool for under-
standing a process, determining whether changes
need to be made, and following the effects of such
changes. It is based on the concept that perfor-
mance differences may be in large part related to
chance variation — random processes — and one
must look for those processes that are "out of con-
trol." Thus, a statistical theory of random varia-
tion was built into a theory of quality manage-
ment. Inherent in the concept is the need for more
information and for theoretically grounded tools
to answer questions (3, 36, 38, 39). The modern
approach to quality must therefore be thoroughly
grounded in both the scientific method and in sta-
tistical theory.

Tools for Statistical Control. The National
Demonstration Project successfully demonstrated
the application of these principles. The purpose of
the quality improvement effort is to eliminate un-
necessary variation. In order to accomplish this,
various graphic approaches have been proposed
and discussed (2, 5, 9, 28, 40); the major graphic
tools are demonstrated in Fig. 1.

The first tool is a flow chart, which repre-
sents steps in any process, thus allowing a clear
and logical understanding of each individual step.
It allows one to zero in at various stages of a pro-
cess in terms of quality, similar to a computer
flow diagram.

A second tool is the cause-and-effect or
Ishikhawa diagram, sometimes called a fishbone
diagram because of its shape. In this diagram, the
effect is the head of the fish and the various po-
tential causes are the bones along the spine. It is
helpful for identifying, displaying, and character-
izing various causes that can lead to a given ef-
fect. It represents relationships and subprocesses,
and thus is an aid to understanding the complex-
ity of problems and determining objectively all
the factors that might cause a given effect. Exam-
ination of this diagram in conjunction with the
flow chart allows one to generate hypotheses
about a given problem in quality. Once these hy-
potheses are generated, one can attempt improve-
ments.

Two additional diagrams, the control chart
and the Pareto bar chart, display information
and data so that conclusions can be readily
drawn. The control chart, developed in the 1920s
by Walter Shewhart and adapted by Deming, is at
the center of this entire process. The chart is
based on the concept that some variation is inher-
ent in any process, and that one must therefore
plot the variation over time of any given param-
eter— for example, the complication rate in a car-
diac catheterization laboratory. These statistical
control charts can demonstrate changes or trends

as alerts that the process may be "out of control"
and requires further investigation.

It is important to distinguish inherent or ran-
dom variation in any parameter from variation
which is real and significant and thus needs to be
addressed. Inherent variation is called "common
cause variation" and abnormal variation, when
the process is "out of control," is called "special
cause variation." The control chart enables one to
determine whether the process is stable and pre-
dictable or contains aberrations. To properly use
these charts, the "normal" degree of variation
must be determined so as to see when the normal
limits are exceeded. Thus, upper and lower con-
trol limits, which can be calculated using statis-
tical procedures such as standard deviations or
can be based on information from other studies,
must be established. The use of control charts is
helpful in the formulating and more important in
the testing of hypotheses to determine if changes
in the process truly affect the data. When there
are special causes, it is important to change the
process itself. However, one must be careful of
what has been termed "tampering": treating com-
mon cause variation as if it were an important
problem for quality and overreacting. The best
medicine against tampering is the use of the con-
trol chart.

The Pareto bar chart arranges various fac-
tors contributing to overall effects in the order of
their decreasing frequency. It is named for the
economist who stated that relatively few citizens
held most of the wealth in an economic system, an
idea applied to quality control by Juran as the
Pareto principle: "Whenever a number of individ-
ual factors contribute to some overall effect, rel-
atively few of those items account for the bulk of
the effect" (28). These diagrams allow one to clearly
see the major confounding factors that enter into a
process that has a potential problem in quality.

A key point is that the methods described
above have been successful both in industry and
in applications to quality in the health care field.
But, in Deming's words, "Industry in America (as
Shewhart said) needs thousands of statistical-
minded engineers, chemists, physicists. Doctors
of Medicine, purchasing agents, managers" (35).
Deming's methods must be applied to the quality
process if we are to succeed; however, blind appli-
cation of statistics to the process will not in itself
contribute to improved quality. A significant
change in overall goals, philosophy, and direction
of the total organization is vital to success; health
care institutions must "organize for quality" (3).

Deming produced a total framework on
which to build this quality endeavor, his 14 points
(35), widely quoted in many settings and applied

Vol. 60 No. 5

QUALITY, DEMING'S PRINCIPLES, AND PHYSICIANS— TEICHHOLZ

353

and adopted in industry and — of particular inter-
est here — in health care settings, especially to the
quality process, as discussed below.

Point 1: Create constancy of purpose for im-
provement of product and service. Constancy of
purpose is extremely important not only for qual-
ity but for the entire organizational structure and
goals of the hospital. Goals must be defined, and
appropriate resources put into place to accom-
plish these goals. Long-term rather than short-
term, put-out-the-fire solutions must be empha-
sized. To accomplish the long-term goals,
innovation and rethinking are required. In terms
of the quality process, one must not look for quick
fixes, but rather recognize and instil quality as
central to the organization. True commitment on
all levels is essential for quality improvement;
this consistency of purpose implies not just filling
out forms for reporting agencies or preparing for
periodic Joint Commission visits, but rather the
adoption of the other Deming principles in or-
der to change the system. Deming urged, "Trans-
late this constancy of purpose into service to pa-
tients and the community. . . . The Board of
Directors must hold on to the purpose" (35).

Point 2: Adopt the new philosophy. The new
approach to management and to quality improve-
ment must be not just a technique but a total
philosophy. The organization must understand
that the best way to improve quality is to improve
the process and do things right the first time.
Both individual and corporate attitudes must
change, and the new philosophy must pervade the
entire organization, including both management
and physicians. The organization must truly be-
lieve in quality. If this is not achieved, the concepts
of continuous quality improvement and the im-
provement of the process will fail. The other points
give us insight into what can be done to aid in the
adoption and acceptance of this new philosophy.

Point 3: Cease dependence on inspection to
achieve quality. A major philosophical concept
should guide the change from quality assurance
to continuous quality improvement. Deming said
that the old way was to "inspect bad quality out"
and the new way was to "build new quality in"
(35). True quality is derived not from inspection
but from improvement of the entire process. The
current quality assurance system is unfortu-
nately the ultimate inspection system, with re-
views and audits and, for the most part, exami-
nation of bad outcomes and affixing of blame.
However, inspection comes too late in the process
of health care to improve quality. When a compli-
cation of a procedure has occurred, a medication
error has been made, or a patient has already
fallen and broken his or her hip, it is too late. The

most effective inspection process will not undo the
damage. Patients deserve better. In many cases, a
proper assessment of the process before the event
could have led to improvement in the process and
avoidance of a poor outcome. Continuous quality
improvement attempts to accomplish this. By
looking prospectively at the process, using appro-
priate statistical techniques and capturing proper
information from all patient encounters, not just
those with poor outcome, one can appropriately
review not just the individuals involved but the
process itself.

Deming never suggested the abolition of in-
spection but rather emphasized not depending on
inspection. Inspection should be an option rather
than the rule. In fact, one needs inspection to
monitor how one is doing, especially when up-
grading quality. One can use control charts (see
above) to pick out special areas for inspection.
The key is that the standard method for improve-
ment is not based on this inspection process. How-
ever, until regulatory agencies are also aware of
and adopt the same philosophy, the filling out of
reports of poor outcomes will unfortunately still
be with us. One challenge for the future is how to
incorporate this appropriately in the continuous
quality improvement process.

Point 4: End the practice of awarding busi-
ness on the basis of price tag alone. Deming am-
plifies point 4 in regard to trying to use single
source vendors. He sees vendors as partners in a
long-term relationship that should lead to higher
quality. How does this apply in the health care
setting? First, it applies directly to conducting the
business of the hospital in the Purchasing De-
partment. A hospital is a business organization,
just as an industrial organization is, and similar
management principles apply. The most success-
ful industries have learned that the total cost of
supplying a service or producing a product is not
just purchase price, whether of raw materials or
other necessities. The hospital must not buy med-
icines or supplies purely on the basis of price. This
principle also applies to decisions about personnel
levels. Although reducing personnel giving direct
patient care in various areas may initially appear
to reduce cost, the overall future costs resulting
from quality problems may be quite dramatic.
One must not look solely at the current price tag
but also at long-term costs for both equipment
and personnel. Deming is succinct: "Price has no
meaning without a measure of the quality being
purchased" (35).

Point 5: Improve constantly and forever ev-
ery process for planning, production, and service
to improve quality. Point 5 follows from Deming's
points 1, 2, and 10 and is the essence of continu-

354

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

ous quality improvement. The process must not
be static. An organization, and the quality of its
products, must be constantly undergoing change
brought about by the appropriate application of
statistical control principles and of continuing
evaluation. The use of statistical control methods
is central to improvement. In addition, quality is
not a one-time effort. The quality process must
not be used to put out fires, but rather to change
overall management and processes. One must not
set a target and then, when it is reached, stop and
feel that the process is complete. Reassessments
and attempts at improvement of performance
must constantly recur. Even if a complication
rate for a procedure is "within local or national
standards," we could still further improve our
performance and the quality of care to our pa-
tients.

How do we accomplish this? By using statis-
tical methods of quality control; however, they
must be applied using the conceptual framework
of the Deming-Shewhart Plan, Do, Check, Act
(PDCA) cycle (9, 19, 35) (Fig. 2). The first step is
to plan, beginning by studying a given process in
detail, using the scientific and statistical meth-
ods. For this study one must know what data are
available and what data need to be obtained. By
the use of flow diagrams and cause and effect
charts, one should determine what possible
changes might improve the process, then create a
plan for making a change and a clear method for
observation of the change to determine its effects
on the total process.

In the second step, the "do" portion, a single
change is made on a small scale. In the third step,
the "check" phase, observations are made and
data obtained on the influence of the change on
the various effects. The fourth phase is to act on
the observations — appropriately examining the
data, testing hypotheses, and coming to conclu-
sions.

Once it has been determined what one has
learned from the change or intervention, then one
must decide whether it was successful and to
what extent. This information should then be
used to move to a newer cycle, either to institute
this change more widely and reexamine its ef-
fects, or to modify the plan and examine an addi-
tional area for improvement of the same process.
Thus by a continuing process we slowly continue
to improve on the quality of our medical care by
climbing to successive cycles, that is, quality im-
provement that is truly continuous. To remove
any connotations of checking, blocking, or reining
in this never-ending process, which aims at a
complete look, Deming recently substituted the
word "study" for the original word "check" in the

third step of the PDCA cycle, which thus is now
known as the PDSA cycle (13, 14).

Point 6: Institute training on the job. Each
individual in the organization must have the ap-
propriate training to know what his or her job is
and how to improve performance. Training has
always been part of the nursing process. We must
also appropriately train technicians and other
members of the health care team, including phy-
sicians. Credentialing of house officers should not
be by the "see one, do one, teach one" process, but
by formal plan with formal evaluation. All mem-
bers of the organization must be trained in the
quality process, including at least an introduction
to the concept of control mechanisms. Supervisors
must be seen as facilitators and must appreciate
that individuals learn in different ways, requir-
ing innovation in training. However, unless uni-
versal training is in the concepts and methods for
improved quality, the entire process is doomed to
fail.

Point 7: Institute leadership. Deming has
said that "the job of management is not supervi-
sion, but leadership" (35). Leadership must sup-
port a pervasive climate of concern for quality
and must maintain and support Deming's princi-
ples, especially constancy of purpose and adopting
the new philosophy. All must work together on
the quality process. Human error does occur; how-
ever, few workers go to work intending to perform
their job poorly. Managers should be like coaches,
helping their workers in the continuous quality
improvement process and in striving for excel-
lence.

Point 8: Drive out fear. Driving out fear is
vital to accomplish the other goals. People must
feel secure in their position and must not be
afraid of making suggestions out of fear of how
their supervisors may react. If people are afraid to
ask questions when they do not understand, the
same problems will be repeated over and over.
Workers must be given the benefit of the doubt
and systemic rather than individual problems
must be emphasized.

Fear is built into our health care system.
Nurses are usually afraid to question doctors.
Physicians fear the consequences of risk alerts or
reports to regulatory agencies. In addition, the
reporting process has an adversarial tone: nurses
file risk alerts and are "responsible" for incident
reporting, while physicians sometimes have the
perception that such reports constitute "telling on
physicians." Another fear of the quality assur-
ance process is of severe disciplinary actions,
when the fault lies primarily with the system. We
must adopt a new attitude removing fear from the
quality process to evince full cooperation and as-

■

Vol. 60 No. 5

QUALITY, DEMING'S PRINCIPLES, AND PHYSICIANS— TEICHHOLZ

355

sistance from all members of the health care
team.

Point 9: Break down barriers between staff
areas. No communication can take place if fear is
endemic. Unfortunately, the hospital medical sys-
tem entails a dichotomy between management
and doctors, between doctors and nurses, and
even between the quality assurance and the risk
management departments. Yet teamwork be-
tween departments, not competition, is a must,
because the entire quality process is full of inter-
connected processes in which individuals are both
providers and consumers. We must develop a to-
tal integrated approach to risk management and
quality improvement, in which all staff work with
common data for common goals. For example, in
patient areas the quality process must be a com-
bined effort of nurses and physicians. Reporting
systems should not be dual or multiple, but to-
tally integrated. Deming has said: "We must
break down the class distinctions between types
of workers within the organization — physicians,
non-physicians, clinical providers versus non-
clinical providers, physician to physician" (35).

Point 10: Eliminate slogans, exhortations,
and targets for the work force. Slogans may be
catchy but are no substitute for a total philosophy
to develop quality improvement in an organiza-
tion. Slogans usually imply that "bad apples" or
noncaring people need to be exhorted. Nor will
slogans used as targets improve quality. How-
ever, slogans may have a limited role in fostering
group identification and teamwork, if the slogans
really come from workers and are part of the total
institutional philosophy and practice.

Point 11: Eliminate numerical quotas for the
work force and numerical goals for managers.
According to Deming, quotas in industry, espe-
cially for quality, are counterproductive. Quality,
not quantity, is important. However, improve-
ment of quality can lead to increased efficiency
and therefore to increased quantity as well. If per-
formance differences are largely due to chance,
setting targets will not alter the process. In addi-
tion, numerical targets can be manipulated — as
in "If you want to hit the target, shoot first and
then call what you hit the target." Using targets,
whether they be 85%, 90%, or 95% compliance
with a given standard, is alien to the concept of
continuous quality improvement. Even if a pro-
cess is deemed good, it can be made better. We
should not work for numerical goals but con-
stantly strive to improve ourselves, the process,
and the organization. Deming has stated: "A
quota is a fortress against improvement of quality
and productivity. I have yet to see a quota that
includes any trace of a system by which to help

DEMING-SHEWHART(PDCA) CYCLE

11^^ ' 1

CHECK

Cycle n

Cycle I

Fig. 2. Deming-Shewhart cycle: Plan (P), Do (D), Check (C),
Act (A).

anyone do a better job. A quota is totally incom-
patible with never-ending improvement" (35).

Point 12: Remove barriers that rob people of
the pride of workmanship, including annual rat-
ing or merit systems. Perhaps one of Deming's
most controversial principles is that an annual
rating and merit system works against the other
general principles because it tends to reward in-
dividual, not group, effort, thus undermining
teamwork, and is based on the idea that individ-
uals rather than systems are the root of problems.
He feels emphasis should be on the sense of a job
well done and the pride of participating in im-
proving the system. He also believes that merit
awards are not necessary if all people are trying
hard to do their best and know what is expected of
them. Remember that even in ideal circum-
stances, with ideal productivity and ideal quality,
half the individuals in a given area will be per-
forming below average and half above average.
Instead of concentrating on merit ratings and
their relation to salaries, the entire work force
must be instilled with the idea that quality is a
proper goal and that people should take pride in
the performance of their job. This may be diffi-
cult, but it is extremely important if the quality
improvement process is to succeed.

Point 13: Institute a vigorous program of ed-
ucation and self improvement. Quality begins
and ends with education. Continued learning for
all in the hospital setting must be the rule. People
should be educated in various ways to improve
their job skills and to advance within the organi-
zation. Unfortunately, education and training
tend to disappear early with hospital budget cuts,
although workers educated in relation to their job
will be more likely to improve job performance.
Equally important, the whole organization — at
the lowest as well as the highest levels — must
be educated in the concepts of quality and, where
appropriate, in the scientific methods and statis-

356

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

tical techniques involved. Everyone must clearly
understand his or her own role in contributing to
the improvement of quality in the organization.

Point 14: Put everyone in the company to
work to accomplish this transformation. This
point encapsulates the previous points. Everyone
in the organization must understand the impor-
tance of quality and must agree on the new phi-
losophy. The major technique for accomplishing
this transformation is not only the philosophical
adoption of these principles but the application of
statistical methods and the Deming-Shewhart
PDCA cycle to continuous quality improvement.

The Role of Physicians

One of the significant reasons for the failure
of the current quality assurance process is the
general lack of physician interest and involve-
ment. Several reasons for this have been re-
counted above. For many physicians, "quality" is
equated with "trouble." Even the peer review pro-
cess primarily examines single episodes of poor
outcome as deviations outside the system. The
premise is that quality can be improved by sin-
gling out those physicians involved in the care of
patients with poor outcomes and assuming that
the physicians have not provided an acceptable
level of care (27).

Physicians must be part of the solution
rather than a significant part of the problem. Ju-
ran has described the problem of an "immune re-
action" to the quality process; physicians may
prove to be "killer lymphocytes" in opposing qual-
ity processes in health care organizations (5). To
succeed, continuous quality improvement re-
quires the participation of physicians on all lev-
els, as Berwick notes: "quality improvement has
little chance of success in health care organiza-
tions without the understanding and participa-
tion, and in many cases, the leadership of individ-
ual doctors" (3).

Is it possible for physicians to become ac-
tively involved in the process? The National Dem-
onstration Project yields some insight. Although
the Project found incorporating physicians into
the planning and team processes was initially dif-
ficult, once physicians became involved they were
active contributors (28), in part due to their fa-
miliarity with the use of scientific methods at the
heart of the statistical control process. The Proj-
ect concludes: "In fact, barriers to physician in-
volvement may turn out to be the most important
single issue impeding the success of quality im-
provement in medical care" (28).

A major method for accomplishing physician
involvement is by education, at all levels. Medical
schools must add to their training curriculum an

understanding of the concepts of continuous qual-
ity improvement as well as an introduction to sta-
tistical control methods and the application of sci-
entific and statistical methods not only to clinical
research and patient care but also to the quality
process. Since a large portion of acute care is pro-
vided by housestaff, especially in teaching insti-
tutions, all housestaff need an educational pro-
gram that crosses departmental boundaries.
However, the most important group to be trained
are the attending physicians themselves; they are
the primary providers of health care both inside
and outside the hospital. Although the emphasis
has been on quality issues inside the hospital,
these principles and concepts must be adopted as
part of an overall philosophy and applied to out-
of-hospital practice as well. In addition, the Board
of Trustees of the institution must also be edu-
cated and have a total commitment to this pro-
cess; this may pose less of a challenge, since many
trustees may already be familiar with the indus-
trial applications of the principles of Deming and
others. The key, however, to the success of con-
tinuous quality improvement rests in the hands
of the chairpersons of the hospital and medical
school departments and, in academic centers, in
the dean and his or her office as well. They must
actively embrace the concept, be its champions,
and actively lead physicians and faculty. This
will require the allocation of adequate resources
for both education and implementation and will
entail initial costs, space, and personnel. How-
ever, we cannot afford not to make this invest-
ment in quality, which will have tremendous pay-
offs in the future.

Furthermore, institutions must break down
dichotomies between the physician hierarchy, the
hospital administration hierarchy, and in many
cases the nursing hierarchy. To be successful in
involving physicians, health care organizations
will have to remove critical barriers and the fear
of surveillance. All members of the team must
work together nonadversarially to improve the
quality of hospital services and patient care.
Quality improvement requires "candid, non-retri-
butional self-examination" (27).

To begin enlisting physicians — and they
must enter the process at an early stage — unre-
alistic expectations of overnight achievement
must be avoided. After the development of a true
institutional commitment, a few clearly defined
projects are necessary to involve physicians and
to demonstrate the usefulness of the technique. If
one starts with small steps and uses statistical
control methods and the Deming principles, in-
cluding repeated Deming-Shewhart PDCA cycles
(Fig. 2), in my opinion the superiority of the ap-

Vol. 60 No. 5

QUALITY, DEMING'S PRINCIPLES, AND PHYSICIANS— TEICHHOLZ

357

proach will quickly become evident. Quality must
be improved in small steps, and initial successes
in clearly circumscribed areas will serve to con-
vince more and more health care providers of the
utility of the process. Actively involving physi-
cians will not be easy, but the results of the Na-
tional Demonstration Project show that it can be
done. Physicians must learn to see how their own
interests can be served by participation and how
their own future is also connected to the health
care organization in which care is given; the pro-
cess must be actively physician driven. This is the
central issue for the future success of continuous
quality improvement.

Conclusions

The words "crisis" and "opportunity" are rep-
resented by the same Chinese ideograph. The cur-
rent crisis of quality in health care in America
contains the opportunity to go forward and solve
many of the problems that exist. Continuous
quality improvement must not be just a slogan.
The entire organizational structure, including
not only administrative leadership but also the
community of physicians, must be committed to
it. Japanese industry and subsequently American
industry have embraced many of Deming's con-
cepts. It is clearly time for the health care indus-
try to do the same. However, these techniques are
not a quick fix for all the ills and problems of
American medicine. We must also recognize that
"quality issues are larger than any particular
spokesperson and supersede passing fads, but to
keep them from becoming cultish, the ideas
rather than the guru must be emphasized" (23).

The techniques outlined in this paper are a
starting point. They will give us the necessary
tools for the future improvement of the quality of
care. In addition, this philosophy should improve
the organization in other ways. The costs of poor
quality — inefficiency, reworking, excess lengths
of stay, malpractice exposure, and litigation — are
among the major problems in the health care sys-
tem. At the same time, fully estimating the ef-
fects of quality improvement on health care costs
is impossible. The industrial quality consultants
in the National Demonstration Project com-
mented that the problems of waste, reworking,
and variation were greater in the health care in-
dustry than in many other industries. They esti-
mated that the cost of poor quality was signifi-
cant; in fact, several suggested that these costs
could approach almost half of the total health
care bill (28).

If the health-care industry is to survive its
current crisis and capitalize on the new opportu-
nities, it must adopt the philosophy and methods

of continuous quality improvement. David
Garvin of the Harvard Business School has con-
cluded (41):

The experience of several participants suggests that qual-
ity improvement, far from being a diversion, leads directly
to greater efficiency and cost reduction. Indeed, if, as seems
increasingly likely, the basis for competition in health care
shifts from a pure price basis to some combination of price,
quality of service, and quality of care, then the issue of
quality will be fundamental to the survival of many health
care organizations. Far from being an interesting, but tan-
gential, activity, there is reason to believe that quality is
likely to become an important basis for competing in health
care over the next decade.

Acknowledgments

I thank Carol Calame and. Marcy Leoncini for their secretari-
al assistance in the preparation of this manuscript.

References

1. Bartlett RC. Trends in quality management. Arch Pathol

Lab Med 1990; 114:1126-1130.

2. Laftel G, Blumenthal D. The case for using industrial

quality management science in health care organiza-
tions. JAMA 1989; 262:2869-2873.

3. Berwick DM. Continuous improvement as an ideal in

health care. N Engl J Med 1989; 320:53-56.

4. Eskildson L, Yates GR. Lessons from industry: revising

organizational structure to improve health care quality
assurance. QRB 1991; 17:38-41.

5. Merry MD. Total quality management for physicians:

translating the new paradigm. QRB 1990; 16:101-105.

6. Mills DH, ed. Report on the Medical Insurance Feasibility

Study. San Francisco: Sutter, 1977.

7. Report of the Harvard Medical Practice Study to the State

of New York. Patients, doctors, and lawyers: medical
injury, malpractice litigation and patient compensation
in New York. Cambridge, MA: Harvard University
Press, 1990.

8. Garvin DA. Managing quality: the strategic and compet-

itive edge. New York: The Free Press, 1988.

9. Walton M. The Deming management method. New York:

Perigee Books, 1986.

10. Walton M. Deming management at work. New York: GP

Putnam's Sons, 1990.

1 1 . Aguayo R. Dr. Deming: the American who taught the Jap-

anese about quality. New York: Lyle Stuart, 1990.

12. Garbor A. The man who discovered quality. New York:

Times Books, 1990.

13. Dobyns L, Crawford-Mason C. Quality or else. Boston,

MA: Houghton Mifflin, 1991.

14. Scherkenbach WW. Deming's road to continual improve-

ment. Knoxville, Tenn: SPC Press, 1991.

15. Scherkenbach WW. The Deming route to quality and pro-

ductivity. Washington, D.C.: CEEP Press Books, 1991.

16. Berwick DM. Quality assurance and measurement prin-

ciples: the perspective from one health maintenance or-
ganization. In: Hughes EFX, ed. Perspectives on qual-
ity in American health care. Washington, D.C.:
McGraw-Hill, 1988.

17. McLaughlin CP, Kaluzny AD. Total quality management

in health: making it work. Health Care Manage Rev
1990; 15:7-14.

18. Jones FG. Continuous quality improvement (CQI): solu-

tion to QA shortcomings. J SC Med Assoc 1990; 86:593-
596.

358

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

19. Milakowich ME. Creating a total quality health care en-

vironment. Health Care Manage Rev 1991; 16:9-20.

20. Marszalek-Gauchar E, Coffey RJ. Transforming health-

care organizations: how to achieve and sustain organi-
zation excellence. San Francisco: Jossey-Bass, 1990.

21. Creps LB, Coffey R, Warner PA, McClatchey KD. Inte-

grating total quality management and quality assur-
ance at the University of Michigan Medical Center.
QRB 1992; 18:250-258.

22. Postal SN. Using the Deming quality improvement

method to manage medical record department product
lines. Top Health Rec Manage 1990; 10:34-40.

23. Albert J, Giligan K, Deevy E. Ready for quality? How one

hospital introduced the Deming methods. Hosp Topics
1990; 68:7-10.

24. Lynn ML, Osborn DP. Deming's quality principles: a

health care application. Hosp Health Serv Adm 1991;
36:111-120.

25. Re R, Kroused-Wood MA. How to use continuous quality

improvement theory and statistical quality control
tools in a multispecialty clinic. QRB 1990; 16:391-392.

26. Hughes JM. Total quality management in a 300-bed com-

munity hospital: the quality improvement process
translated to health care. QRB 1992; 18:293-300.

27. Kritchevsky SB, Simmons BP. Continuous quality im-

provement: concepts and applications for physician
care. JAMA 1991; 266:1817-1823.

28. Berwick DM, Godfrey AB, Roessner J, eds. Curing health

care: new strategies for quality improvement. San
Francisco: Jossey-Bass, 1990.

29. Roberts JS, Schgue PM. From QA to QI: the views and

role of the Joint Commission. Qual Lett 1990; May.

30. O'Leary D. Future of TQM for the Joint Commission on

Accreditation of Healthcare Organizations. In: Melum

MM, Sinioris MK. Total quality management: the
health care pioneers. Chicago, 111: American Hospital
Association Publishing, 1992.

31. Eubanks P, Grayson M. The CEO experience: TQM/CQI.

Hospitals 1992; 66(ll):24-36.

32. Leebow W, Ersoz CJ. The health care manager's guide to

continuous quality improvement. Chicago, 111: Ameri-
can Hospital Association Publishing, 1991.

33. Leebow W. The quality quest: a briefing for health care

professionals. Chicago, 111: American Hospital Associa-
tion Publishing, 1991.

34. Melum MM, Sinioris MK. Total quality management: The

health care pioneers. Chicago, 111: American Hospital
Association Publishing, 1992.

35. Deming WE. Out of the crisis. Cambridge, MA: Massa-

chusetts Institute of Technology Center For Advanced
Engineering Study, 1986.

36. Gillem TR. Deming's 14 points and hospital quality: re-

sponding to the consumer's demand for the best value
health care. J Nurs Qual Assur 1988; 2:70-78.

37. Thompson RE. Some practical applications of Deming's 14

points. J Quality Assur 1990; 12:22-23.

38. Neuhauser D. The quality of medical care and the 14

points of Edwards Deming. Health Matrix 1988; 6:7-10.

39. Darr K. Applying the Deming method in hospitals: parts

1 & 2. Hosp Topics 1989; 67:4-5, 1990; 68:4-6.

40. Kinlaw DC. Continuous improvement and measurement

for total quality: a team-based approach. San Diego,
CA: Pfeiffer, 1992.

41. Garvin D. Reflections on the future. In: Berwick DM, God-

frey AB, Roessner J, eds. Curing health care: new strat-
egies for quality improvement. San Francisco: Jossey-
Bass, 1990, 165.

Risk Management in Health Care:

Where Did It Come From and Where Is It Going?

Barbara Challan, R.N., A.R.M., Jane McConnell, R.N., J.D., A.R.M., and Alice Walsh, R.N.

Abstract

This article reviews some of the historical developments contributing to the evolution of
risk management and its current role in health care. The scope of activities and concerns
that today's health care risk managers address and the need for involvement by the entire
health care team are discussed.

When a physician who is innocently going about
the daily business of seeing patients at the office
is told the hospital risk manager is on the phone,
he or she knows, even before picking up the
phone, that there is a problem. Only a few years
ago, many physicians had never even heard of
risk management. Now it is hard to avoid the
constant influx of bulletins, newsletters, and sem-
inar notices hawking the latest in how one may
end up in court. Where did this relative upstart of
a health care discipline come from, and where is it
going?

Historical Background

Until the 1970s, the health care field was less
affected by the costs of liability losses than many
other industries in this country. Furthermore,
until then, most businesses were more concerned
about property losses than liability losses. The
liability revolution had not yet struck. For health
care, the patient rights movement of that decade
abruptly changed all that. Patients no longer
wanted to be "done to," but instead demanded to
participate, to be informed before giving consent
to treatment, and to sue for damages when they
believed negligent care had been provided.

The courts supported this new patient activ-
ism, awarding huge financial compensation for

From the Risk Management Department, FOJP Service Cor-
poration, 130 East 59th Street, New York, N.Y. 10022. Ad-
dress reprint requests to the first author, now Director of Cor-
porate Communications, at that address.

The Mount Sinai Journal of Medicine Vol. 60 No. 5 October 1993

injuries in medical malpractice suits. This funda-
mental change in the attitude of patients and the
courts, along with shifting public policy, tore
through the buffers of doctor-patient trust and
professional insularity to make hospitals and doc-
tors targets for unprecedented liability losses.

At the same time, hospitals in many parts of
the country were losing the charitable immunity
from civil suits that had long protected their op-
erating budgets and other assets. Not-for-profit
institutions became as vulnerable to losses from
lawsuits as for-profits. Furthermore, plaintiffs'
attorneys were becoming increasingly skillful at
trying medical malpractice cases, causing both
hospitals and physicians to fear them more. Many
industries, including health care, lobbied state
legislatures for tort reforms to help stem the over-
whelming losses from liability claims.

The 1980s brought cost containment initia-
tives and more (and more) governmental regula-
tions. Political scrutiny was joined by increas-
ingly aggressive media coverage; new, exotic
medical technologies and medical mishaps pro-
vided plenty of grist for the reporter's mill. With
health care delivery now on the priority list of
national problems, the field will continue to op-
erate under the public's gaze. The quest for civil
tort reforms will continue as well, because of
ever-expanding theories of liability.

In the last two decades, the burden of liabil-
ity losses, as well as compliance with federal and
state statutes, has become an increasing drain on
financial resources used for patient care, re-
search, investment in new equipment, and other

359

360

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

improvements in delivering medical service. To
stop this liability drain, the health industry be-
gan to allocate more resources to risk manage-
ment in an attempt to control risks more effec-
tively while improving the quality of care.

Until the late 1970s, large hospitals had at
most an insurance manager who purchased the
requisite insurance coverage and tracked the
sources of some insurance losses. When a hospital
realized it could no longer afford to just "roll with
the punches," and the public continued to demand
more oversight, a risk manager was hired and
charged with gaining control over liability losses
through prevention and quality improvement.
Smaller hospitals assigned risk management as a
part-time function to an administrator.

Evolution of Risk Management

To health professionals who are schooled in
keeping patient information private, the whole
process of risk management can be uncomfortably
candid — not to mention suspiciously trendy. How-
ever, risk management is not a recent phenome-
non and has its roots in insurance. Risk manage-
ment has had a well-established role in other
industries for hundreds of years and in medicine
for at least a century. Gathering and analyzing
information, while safeguarding it from improper
disclosure, is essential to managing risks and
achieving loss reduction.

Since the dawn of the industrial age, people
have been negotiating ways of protecting busi-
ness owners from avoidable losses. The roots of
insurance can be traced back 300 years, to the era
of hazardous overseas trade, when Lloyd's of Lon-
don first gathered small sums of money from ship
owners to cover the cost of lost cargo.

At that time, "risk management" efforts were
focused on financial losses from sunken cargo.
Over time, both insurers and business owners
perceived the benefit of prevention through fire
safety, machine safety, protection from natural
perils, ad infinitum. In addition to the intrinsic
worth of human life and health, the increasingly
sophisticated skills possessed by workers came to
be viewed as yet another kind of potential loss. A
loosely organized, often totally unorganized array
of many types of insurance and mandatory and
voluntary loss prevention mechanisms developed
to protect individuals and organizations from
harm and, ultimately, financial loss. Not until
the industrial revolution, when work-related in-
juries grew exponentially, did personal injury li-
ability claims become a major issue; management
of the risk of loss eventually became integral to
determining an organization's viability and po-
tential for growth.

Risk management now covers the gamut of
industrial concerns from plant safety to product
reliability. As American businesses become more
complex, the potential for liability loss multiplies,
and risk management becomes even more criti-
cal. Unexpected exposures to loss continuously
arise: traces of benzene are found in Perrier, one
of the most popular bottled waters imported into
this country; someone becomes ill from Tylenol
that has been laced with poison; shards of glass
are found in a bottle of baby food. These are just a
few of the well-known cases in which the risk
manager is called in to use all of a company's
resources — legal, technical, research and develop-
ment, public relations — to reduce potential liabil-
ity losses. The extent to which an organization
manages such crises well can be a matter of sur-
vival. The Tylenol case, for example, has become
a classic case study in good crisis management
and good risk management, in this case to protect
against loss of public confidence and thus loss of
revenue. All industries — food, transportation,
banking, manufacturing, entertainment, all
manner of services — have risk managers today.
In health care, especially, risk managers play a
crucial role in crisis damage control.

Risk Management in
Health Care

The health care industry has become enor-
mously complex and continues to evolve techno-
logically. Risk managers are often viewed nar-
rowly as the people who review malpractice
claims and awards. However, the purview of risk
management in today's medical facility is much
broader. All potential risks of loss can fall under
the umbrella of the risk management office.

If a student is robbed in a medical center
housing complex, the risk manager works with
security personnel, law enforcement officials, at-
torneys, and student affairs and public relations
departments to respond effectively to the incident
and to develop preventive measures for the fu-
ture.

Risk managers work with insurance brokers
or insurance services firms to analyze insurance
coverages and ensure that appropriate coverage
is in place. In modern health care facilities, the
different types of insurance required can number
over twenty, of which medical malpractice is only
one, albeit an extremely important one.

For emerging technologies, such as infertility
treatment, risk managers must keep pace with
new theories of liability that may add strange
new twists to necessary preventive measures.
Similarly, the continuous promulgation of gov-
ernmental regulations has expanded the regula-

Vol. 60 No. 5

RISK MANAGEMENT— CHALLAN ET AL.

361

tory-compliance area of risk management, to pre-
vent avoidable losses from fines and damage to
the hospital's reputation.

Finally, when a process involved in deliver-
ing health care services is associated with medi-
cal malpractice claims and compensatory dam-
ages, risk managers must work with everyone
involved with that process — physicians, nurses,
physician assistants, technicians, administrators,
manufacturers of medical products, and so on — to
find safer ways to deliver the care. This includes
investigating individual incidents to detect spe-
cific weaknesses in the system and looking for
overall patterns that demonstrate the need for
systemic change. Ultimately, risk management
involves any activity in an organization that
can damage its ability to deliver services safely,
its reputation with the public, or its fiscal sol-
vency.

Essential Role for All Clinical, Managerial,
and Support Staff. Given the scope of information
with which risk managers need to be familiar, it
is imperative for them to draw on technical ex-
pertise in the health care organization. An effec-
tive risk manager sits at the hub of a health care
facility's information sources, gathering data that
enables him or her to recognize and avoid poten-
tially hazardous situations, evaluating actual
and potential claims rapidly to stem financial
loss, and calling on expert resources for advice.

The success of a risk management program
depends above all on the availability of a network
of experts for guidance in developing risk-reduc-
ing programs. As a generalist, the risk manager
can recognize areas of trouble and make sugges-
tions to reduce hazards, but he or she needs the
input of physicians and other clinical and non-
clinical personnel to develop procedures that
function effectively in the medical arena. Without
input from clinicians, management, and support
staff who work in the intensive-care units, the
emergency rooms, the delivery rooms, the operat-
ing rooms, indeed everywhere medicine is prac-
ticed, the risk manager is hampered in designing
new risk control procedures that are technically
feasible, practical, and effective.

Furthermore, the risk manager needs the ex-
pertise of clinicians and other front-line person-
nel to figure out which battles are worth fighting.
Zero risk is impossible to attain. Some level of
exposure to risk, often a high level, is a reality of
the health profession. The most critical job of the
risk manager is to determine which potential haz-
ards should have priority, given the limitations of
funds, time, and staff available at an institution.

For risks directly associated with invasive
procedures or the use of certain technologies, phy-

sicians can be key in evaluating where risk man-
agement will functionally cut down on financial
loss associated with medical care itself; they
know where the most injurious errors are com-
mon as well as what alternative protocols might
reduce injury. Physicians can also be helpful in
discouraging the use of risk-management re-
sources in areas where they are least likely to be
effective.

No more or less important than physicians
are other clinical and nonclinical personnel who
can provide information and analysis that bear on
how injuries occur in a particular context. Nurses
are often in the best position to observe the inter-
action between medical procedures and the sys-
tem within which the procedures take place.
Technicians who routinely use certain types of
equipment may be the first to notice minor de-
fects that can put patients at increased risk.
Housekeeping personnel may be more likely to
recognize subtle safety hazards on patient care
units. Ideally, everyone in the stream of health
care delivery is involved in continuous evaluation
of the efficacy of services. The current trend in
health care toward implementing principles of
continuous quality improvement, which require
the active involvement of all levels of staff, is con-
gruent with the goals of risk management.

Cooperation between risk management and
health care professionals can be extremely fruit-
ful. Ongoing shared data projects have been un-
dertaken by large companies and groups that pro-
vide medical malpractice insurance and risk
management services. Specialty societies, nota-
bly the American Society of Anesthesiology, have
also pooled data on malpractice claims to develop
a series of safety guidelines to reduce injuries.
Such efforts alert all organizations to malpractice
claim trends either previously not appreciated or
not detectable in individual organizations or
small groups of organizations. The UJA-Federa-
tion hospitals (located in New York City),
through FOJP Service Corporation, have worked
together to develop safety guidelines for anesthe-
siology and are currently working on guidelines
for obstetrics.

In the decades to come, such cooperative ef-
forts will be ever more necessary as advances in
biotechnology cloud even the most basic assump-
tions about life and death and make the medical
profession more vulnerable to unforeseeable lia-
bility losses. Already, previously unimaginable
legal battles are being fought over which divorc-
ing spouse "owns" the fertilized eggs stored at
their local medical center. Other areas that are or
are likely to become grounds for litigation in-
clude:

362

THE MOUNT SINAI JOURNAL OF MEDICINE

• Decisions on which patients will receive medi-
cal treatment available only in limited supply,
such as organ donation

• Genetic research and treatment

• Issues related to HIV transmission

• Issues related to death and dying

The future of health care in this country de-
pends not only on governmental regulations and
other external forces, but on the ability of health
care organizations to direct their finances out of
the courts and into the laboratories and hospitals,
where the benefits will reach the greatest number
of people.

External forces create pressure on the health
care system to change, but are not the actual
agents of change. In order to stop the current
drain of resources to liability claims, risk manag-
ers, physicians, other clinicians, and administra-
tors need to work closely together to foresee and
avoid preventable losses. The health care risk

manager is an ally with common concerns, and
ongoing communication is a positive investment
toward improving the quality of health care.

Sponsored by a grant from Cellcor.

46. Jumping Translocations of Trisomic Long Arms of Chromosome 1
in Acute Myeloid Leukemia (AMD. B. Hauschildt, A. Scalise, L. Sil-
verman, A. Gattani, E. Ambinder, V. Najfeld. Tumor Cytogenetics
Laboratory, Polly Annenberg Levy Hematology Center and Depart-
ment of Neoplastic Disease.

Terminal or telomeric associations (TA) of a trisomic chromosome with
different intact receptor chromosome in a form of jumping transloca-
tions are an unusual phenomena in leukemia. We now report two pa-
tients with history of myelodysplastic syndrome who transformed into
AML and whose leukemic cells showed trisomy of the long arms (q) of
chromosome 1 in jumping translocations with four different receptor
chromosomes.

In patient 1, 74 bone marrow metaphases were examined; 70% had
the trisomic Iq attached to the short arms of chromosome 22 in the
form of der(22)t(l;22)(ql2;pl3); 17.5% of cells had trisomy Iq translo-
cated onto the short arms of chromosome 15: der(15)t(l;15)(ql2;pll)
and 4% of cells had TA between trisomy Iq and the end of the short
arms of chromosome 8: der(8)t(l;8)(ql2;p21). A similar distribution fre-
quency of trisomy Iq in TA with chromosomes 8, 15 and 22 was dem-
onstrated in unstimulated peripheral blood cells: 1%, 14% and 73.5%,
respectively.

In patient 2, the major cell population in bone marrow, detected in
49 cells, had a 47,XY,t(4;ll)(qlO;pl5), -I- 11 karyotype. An additional
rearrangement consisting of trisomy Iq translocated to the short arms
of chromosome 22: der (22)t(l;22)(ql2;pl3) was detected in 12% of cells.
Investigation of chromosomes obtained from unstimulated peripheral
blood cultures revealed trisomy Iq in a TA with chromosome 21:
der(21)t(l;21)(ql2;pl3) (33% of cells) as well as with chromosome 22
(33% of cells). Therefore, blood leukemic blasts had trisomy of the long
arms of chromosome 1 in jumping translocations with chromosomes
21 and 22.

The mechanisms of jumping translocations and their relationship
to the pathogenesis of AML remains speculative because this phenom-

426

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

ena has rarely been documented. Based on the hypothesis that loss of
telomeric DNA repeats, beyond the critical point, may result in chro-
mosomal instability and cell senescence, jumping translocations in leu-
kemic blasts may be explained by the rapid decrease of telomere length
with each cell division. Consequently, some chromosomes may achieve
this critical level in a shorter time. We, therefore, performed fluores-
cence in situ hybridization experiments using telomeric terminal re-
peat DNA probe. Preliminary results indicated a decrease of telomere
repeats on chromosomes involved in these translocations. Jumping
translocations were detected just before each patient died and it is
conceivable that these translocations are the result of telomere loss and
may represent markers for cell death.

47. Cardiotoxicity of Adriamycin in Multidrug Resistant Transgenic
Mice. Luis Isola,* Steven Frank,* Pablo Cagnoni,* Jon Gordon. t *Di-
vision of Hematology, Department of Medicine, tDepartment of
OBS/GYN and Reproductive Science, Mount Sinai Medical School,
NY, NY.

Adriamycin (ADR) is one of the most widely used drugs in the treat-
ment of hematologic malignancies. With the use of colony stimulating
factors and bone marrow rescue the severe myelosuppression associ-
ated with this agent can be substantially alleviated. However, cardio-
toxicity is still a dose limiting toxicity. This limits the use of anthra-
cyclins in salvage and bone marrow transplant preparative regimens.
A general strategy proposed to enhance the safety margin of chemo-
therapy and allow further dose escalation in the transfer of drug resis-
tance genes into cells that are target for toxicity. Among these se-
quences, the multiple drug resistance (MDR) gene encoding the
P-glycoprotein, has been introduced into cells to confer ADR resistance.
Transgenic mice (TM) have been previously used by us and others to
investigate the potential of genetic engineering of bone marrow resis-
tance in vivo.

We have produced three lines of TM with a Herpes virus thymidine
kin£ise (TK) promoter-MDR cDNA construct. Using this model we con-
ducted a pilot study in which we took four animals, 2 TM and 2 con-
trols, through two weekly cycles of intraperitoneal ADR at a dose of 8.5
mg/kg.

In this first study the two TM survived and the controls died one
week after the last injection (difference NS). A larger study included 11
TM and 12 controls which received four cycles of ADR at the same dose,
the survival for the TM was 36% (4/11) and for the controls 42% (5/12)
(difference NS). There was also no significant difference in blood counts
between surviving animals in the two groups at the end of the study.
The examination of the hearts by optical microscopy showed that the
controls had more severe effects of ADR toxicity (myocyte vacuolization
and sarcotubular dilatation).

We conclude that: 1 ) In spite of its ability to elicit high levels of
transcription of other genes in TM, TK is not an efficient promoter to
drive MDR expression in bone marrow precursors, 2) While TM carry-
ingTK-MDR constructs are not a good model for in vivo genetic engi-
neering of bone marrow resistance, they may be useful to investigate
the use of MDR genes in overcoming ADR cardiotoxicity, 3) With an
appropriate method of gene delivery, MDR may be used to protect the
myocardium fi-om ADR toxicity so as to facilitate dose escalation in
regimens containing ADR or in patients who have reached the maxi-
mum cumulative dose currently recommended. A study with a larger
number of mice, P-glycoprotein aissays and electron microscopic assess-
ment of ADR myocardial toxicity is ongoing.

48. High Purity vs Intermediate Purity Factor VIII in HIV-H Hemo-
philiacs: Conclusions of a Prospective 3- Year Study. S. Seremetis, LM
Aledort, TS Lau, H Sacks and the Monoclonal Study Group. The De-
partment of Medicine, Divisions of Hematology and Infectious Dis-
eases, Eind Department of Biomathematics at the Mount Sinai School of
Medicine, New York, NY.

Low CD4 lymphocyte counts are associated with increased risk of pro-
gression to AIDS in HIV infected persons. In order to test the hypoth-
esis that the use of high-purity products in HIV + hemophiliacs would
result in a difference in the rate of deterioration of immune function,
we designed a protocol for a multicenter, prospective, controlled study
of asymptomatic HIV infected subjects with hemophilia A who were
randomly assigned to receive either an intermediate purity or mono-
clonal antibody-purified product.

At study entry, all subjects had CD4 counts «600 and slOO, were
negative for HBsAg, and had not received any antiretroviral or im-
muno-modulating drugs. 60 subjects were recruited and 30 subjects
were assigned to each group. Subjects were permitted to receive an-

tiretroviral therapy during the study. 35 subjects completed the three-
year study, 20 in the monoclonal arm and 15 in the intermediate purity
arm.

There were no differences between the two groups in the occur-
rence of AIDS-defining diagnoses ( 1 patient in each group). There were
however significant differences between the two groups in terms of
changes in absolute CD4 counts.

Time Intermediate Monoclonal p

(Mean ± SD)

0 378.3 ± 126.5 416.3 ± 139.7 0.4130

12 mo. 337.1 ± 116.8 400.3 ± 161.5 0.2088

24 mo. 268.1 ± 126.8 445.9 ± 191.1 0.0037

36 mo. 190.1 ± 114.7 393.3 ± 200.3 0.0007

The group receiving monoclonal antibody-purified concentrates had
essentially stable CD4 counts while a significant drop in CD4 counts
was observed in the group receiving intermediate purity concentrates.
These differences were independent of the use of antiretroviral ther-
apy.

These data, from the largest and longest such study reported, in-
dicate that the use of monoclonal-antibody purified concentrates is as-
sociated with better preservation of CD4 lymphocytes in asymptomatic
HIV + patients with hemophilia A. We also conclude that repeated
exposure to some component of intermediate purity Factor VIII con-
centrates represents a cofactor in the progression of HIV disease in
patients with hemophilia A.

49. Nasal Spray Desmopressin (DDAVP) Experience in Home Care
and Surgical Prophylaxis. S. V. Seremetis and L. M. Aledort. Division
of Hematology, Department of Medicine, Mount Sinai School of Medi-
cine, New York, NY.

The use of plasma products to treat patients with von Willebrand Dis-
ease (vWD) and hemophilia has led to transfusion transmitted dis-
eases. The introduction of DDAVP to the U.S. in 1983 has obviated the
need for plasma for many of these patients. Currently, DDAVP is ad-
ministered by intravenous (IV) and subcutaneous (SO routes. We have
previously reported that intranasal (IN) administration of DDAVP via
a simple atomizer has produced a bioavailability equivalent to an IV
dose of 0.2 ug/ml. In both mild hemophiliacs and vWd patients, this
resulted in elevations in FVIILC, vWF, and vWAg that were compa-
rable to the results of IV administration. Patients whose baseline test-
ing showed favorable results were given the nasal spray for home-care
use for bleeding episodes and prior to minor surgical procedures.

We now have efficacy data from 184 such episodes in 18 patients,
13 with vWD, 3 mild hemophiliacs and 2 hemophilia carriers; patients
self-scored responses to treatment with IN-DDAVP. 10 women with
vWd treated a total of 101 episodes of menorrhagia: 85 episodes were
scored as excellent responses, 9 were scored as good, the remaining 7
were scored as minimal. Also in patients with vWd, 54 episodes of joint
bleeding, 15 of epistaxis, and one laceration were treated with excel-
lent hemostasis. In addition, one patient underwent extensive dental
work and another patient had carpal tunnel relesise; in both cases
patients reported excellent hemostasis. In mild hemophiliacs, 13 epi-
sodes of hemarthrosis were treated with IN-DDAVP; all episodes were
scored as excellent or good responses. One of the hemophiliacs under-
went arthroscopic surgery; hemostasis and healing were judged excel-
lent. Two surgical procedures were performed in hemophilia carriers;
both episodes were scored as excellent hemostasis.

We conclude that IN-DDAVP is an efficacious, simple, convenient
method for the treatment of mild hemophilia and vWD. It is adaptable
for home use and will be a useful addition to the treatment armamen-
tarium for these hereditary coagulopathies.

50. Endothelial Cell Transglutaminase at the Interface of Thrombosis
and Fibrinolysis. X. Zhang, W. Borth, P. C. Harpel. Division of Hema-
tology, Mount Sinai School of Medicine, N.Y.

There is increasing evidence that cellular transglutaminases (TGases)
play a significant role in cellular adhesion and apoptosis (programmed
cell death). TGases are a family of Ca-dependent enzymes which cross-
link some protein substrates. Endothelial TGase is thought to play a
role in atherogenesis that remains to be defined. We have recently
proposed that endothelial TGase functions at the interface of thrombo-
sis, fibrinolysis and tissue modeling by modifying some proteins that
have critical functions in thrombosis and fibrinolysis. Modification of

Vol. 60 No. 5

ABSTRACTS

427

these proteins by TGase either creates a prothombotic or fibrinolytic
milieu. We have recently shown that apolipoprotein (a) (apo(a)), a
unique protein moiety of the atherogenic lipoprotein particle lipopro-
tein (a) (Lp(a)) is a substrate for TGases. This suggested to us that Lp(a)
may be immobilized in the vessel wall upon cross-linking to connective
tissue matrix components. Apo(a) is extensively homologous to plas-
minogen and both are members of a family called kringle proteins
which evolved from an ancestral serine proteinase.

Using probes that identify TGase-reactive glutamines and lysines
we observed that plasminogen is an excellent TGase substrate. Plas-
minogen could also be cross-linked to fibronectin, a well characterized
adhesion protein and TGase substrate. Human umbilical vein endothe-
lial cells (HUVECs) are a source of TGase in vitro. If HUVECs were
incubated with radiolabeled plasminogen, large molecular weight ag-
gregates formed that resisted to reduction and SDS treatment.

These aggregates remained cell associated and no cross-linking of
plasminogen was seen in the culture supematants suggesting that cell
surface associated TGase cross-linked plasminogen to cell membrane
associated proteins. It is possible that plasminogen becomes cross-
linked to pericellular connective tissue matrix components including
fibronectin and proteoglycans. On the other hand, plasminogen might
become cross-linked to a cell membrane protein or "receptor." HUVEC
membranes were prepared by sequential ultracentrifugation steps. The
possibility was addressed whether these membrane preparations con-
tained proteins that showed substrate characteristics for TGase. When
probed for reactive glutamines we found one protein, p 55, that was
highly susceptible for TGase. When probed for reactive lysines we ob-
served one protein that preferentially labeled at a molecular weight of
93 kD.

Our observations indicate that plasma proteins that participate in
thrombogenic (i.e., Lp(a)) or fibrinoljrtic (i.e., plasminogen) processes
are modifyed by TGases. Endothelial cells which are binding sites for
these proteins express also TGase substrates on their cell surface. It is
conceivable that TGase catalyse a tight interaction of these hemostatic
factors thereby Eiffecting the biological characteristics of these proteins.

Infectious Diseases

51. Epithelial Cell Invasion by Mycobacterium avium Complex
(MAC). S. T. Brown,' R. Richardson,^ and L. W. Riley.^ Infectious Dis-
ease Section, Bronx VAMC & Mount Sinai School of Medicine^ and
Department of International Medicine, Cornell University Medical
College.2

The gastrointestinal tract is the most likely portal of entry leading to
infection with Mycobacterium avium complex (MAC) in patients with
AIDS. We examined the interaction between well-characterized clini-
cal isolates of MAC and HeLa cells, a culture line having epithelial cell
characteristics.

Isolates of MAC from AIDS and non-AIDS patients were incubated
for 48-72 hours at a ratio of 100:1 with HeLa cells.

After thorough rinsing, Kinyoun stains of methanol-fixed cultures
showed large numbers of acid-fast bacilli (AFB) associated with HeLa
cells from some strains of MAC while the interaction with other strains
was negligible. Acid-fastness of HeLa cell-associated MAC from some
strains was lost after methanol fixation but present with glutaralde-
hyde fixation. Electron microscopy of HeLa cell-associated mycobacte-
ria showed that they were intracellular and were surrounded by an
electron-transparent zone characteristic of pathogenic mycobacteria.

These findings suggest that there are strain-dependent differences
in the ability of MAC to invade HeLa cells. The cultivation of MAC
with HeLa cells may be a useful model in which to study virulence
factors for MAC.

52. Complications of Adjuvant Steroid Therapy for Presumed Pneu-
mocystis carina Pneumonia (PCP) in an Inner-City Population. T.

Cheung,* F. Wallach, S. Cohen, and H. Sacks. Mount Sinai Medical
Center, New York, N.Y.

Objective. To evaluate the short- and long-term complications of adju-
vant steroid therapy for HIV-infected persons admitted with a diagno-
sis of presumed PCP.

Methods. Retrospective chart review of all HIV-infected persons
admitted with a diagnosis of presumed PCP and on adjuvant steroid
therapy from June to December 1992.

Results. Twenty-one patients have been treated by this method.
Racial groups included 10 Blacks and 11 Hispanics. Their ages ranged
from 25 to 62 (mean 36.1). Their risk groups were: 15 intravenous drug
users (IVDU), 4 heterosexual and two with both IVDU and heterosex-
ual risk factors. Mean CD4 count was 28.4 ± O.e/mm''. PCP therapy
included: 17 received trimethoprim-sulfa and 4 pentamidine; one was
given a subsequent course of primaquine and clindamycin. Average
steroid use was 563.5 mg of prednisone equivalent. Short-term compli-
cations (<1 month) included glucose intolerance in 2, psychosis in 1,
fulminant tuberculosis in 2, disseminated Mycobacterium avium com-
plex in 2. Long-term complications: Cytomegalovirus retinitis in 2 (one
vnth disseminated CMV disease of multiple organs). Two patients died
during therapy. One died two months later of end-stage HIV disease;
two patients were lost to follow-up and the rest were still alive.

Conclusion. Adjuvant steroid therapy for presumed PCP is gener-
ally safe and effective. However, in areas of high tuberculosis inci-
dence, the safety of adjuvant steroid therapy may need to be reevalu-
ated. Diagnostic procedures should be performed to rule out other
possible infections.

53. Prevalence of Latex Glove Allergy in a Population of Health Care
Workers (Operating Room Labor and Delivery Nurses and Phlebot-
omy Workers) at Elmhurst Hospital Center: 1993. Salim Gopalani.
Department of Medicine. Elmhurst Hospital Center.

Latex glove allergy is increasingly recognized as a cause of occupa-
tional morbidity especially among health care workers, who have in-
creased latex exposure in recent years. A study was conducted to assess
the prevalence of latex glove allergy in operating room, emergency
room, labor and delivery nurses and phlebotomy workers at Elmhurst
Hospital Center.

A questionnaire was distributed to 114 health care workers to
identify those individuals who had symptoms of allergy to latex gloves
and also to characterize associated factors such as length of employ-
ment, duration and frequency of use.

The response rate to the survey was 63.1% (n = 72) there were 21
(29.1%) responses indicating latex glove allergy; there were 0% re-
ported allergy to nonoccupational exposure to latex. Seven workers
(9.7%) reported access to non-latex gloves was inadequate.

The prevalence to allergy may be lower, as workers with allergic
symptoms reported promptly as compared to workers (non-responder)
with no symptoms. Latex allergy, as ascertained by self-report in this
study, appears to have substantial prevalence in these workers, and
non-latex gloves should be readily available for workers allergic to
latex.

54. Essential Hypertension in Caribbean Hispanics: Sodium, Renin,
and Effects of Therapy. C. Laffer and F. Elijovich. Division of General
Internal Medicine, Mount Sinai School of Medicine, New York, NY.

Seventy-one Hispanic hypertensives of Caribbean descent (77% Puerto
Rican, 80% female) participated in 116 treatment trials. Age was 57 ±
1 yrs (29-79); diabetes and obesity were present in 27% and 70%,
respectively.

After 3-4 weeks ofi' therapy, UNaV was 143 ± 11 mEq/d (34-445,
n = 66) and exceeded 100 in 64%. PRA was 0.9 ± 0.1 ngAI/mL/hr
(0-3.4, n = 84) and was below the reported lower 95% confidence limit
for PRA/UnaV of normotensives in 64% . PRA correlated with UNaV (r
= -0.36, p < 0.01) but not with age. Double-blind therapy (8-12 wks)
was not given if diastolic BP was <90mmHg(n = ll)or>110(n = 10)
at the end of the washout phase. Randomization to placebo (n = 14)
reduced BP by - 6 ± 4/ - 7 ± 2.

Significantly greater reductions in BP were observed with hydro-
chlorothiazide (H, -23 ± 2/- 11 ± 1, n = 16) and its combinations
with converting enzyme inhibitors (H-ACE, - 23 ± 3/ - 14 ± 2, n = 13)
and p blockers (H-p, - 30 ± 4/ - 17 ± 2, n = 14). H-p was given in
random sequential fashion, permitting calculation that 67% of the BP
fall was due to H. In contrast to the antihypertensive effects of diuretic-
containing regimens, those of monotherapy with ACE ( - 3 ± 4/ - 5 ± 2,
n = 25) or p ( - 8 ± 5/ - 8 ± 2, n = 13) were not different from placebo,
despite reduction in heart rate with p ( - 15 ± 2 bpm, p < 0.001). The
reduction in BP by H correlated with PRA (r = -0.45, p < 0.05) and
was associated with weight loss ( - 2 ± 1 lbs, p < 0.01). Treatment with
H, ACE and H-ACE increased PRA by 0.4 ± 0.2, 0.6 ± 0.2 and 1.8 ±
0.7, respectively (p < 0.05 for each).

We conclude that Hispanic hypertensives are obese with low, but
stimulatable, PRA and high UNaV. We speculate that high sodium
intake is a cultural carry-over of ancestral adaptation to tropical cli-

428

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

mates in Caribbean Hispanics, and that it explains their previously
unreported pattern of response to antihypertensive therapy.

55. Impact of Expanded Tuberculosis (TB) Isolation Policy on a New
York City Urban Teaching Hospital. M. H. Mendelson,* A. Moore, M.
Vigo, L. Finkelstein, B. R. Meyers, J. Dong, J. Solomon, and S. Z.
Hirschman. The Mount Sinai Medical Center (MSMC) New York, New
York.*

Due to the recent epidemic of TB, atypical presentations of TB and
reports of nosocomial transmission, hospitals have implemented ex-
panded indications for TB isolation. At the MSMC these have included
clinical/radiographic respiratory processes in all HIV positive/high risk
patients in addition to classical TB presentations in all patients.

145 patient admissions (pt adm) (139 pts) from 12/1/92 to 2/16/93
were placed on TB isolation. 30 pt adm had a final diagnosis of TB
(20.4%), 28.3% bacterial infection, 13.8% Pneumocystis pneumonia,
33.8% unknown, 5.5% other.

A comparison of TB vs. non TB pt adm showed: mean age (36, 41
yrs.), HIV positive (46.7%, 55.7%), previous TB history of (h/o) (33.3%,
9.6%; p < .01), productive cough (57%, 64%); CXR findings; focal/alve-
olar infiltrate(s) (67%, 50%), cavitation (33%, 10%; p < .01), increased
interstitial markings/infiltrates (3%, 27%; p < .01), upper lobe finding
(17%, 14%), adenopathy (23%, 13%), negative (10%, 12%); and mean
isolation days (26, 10). 30% TB pt adm. were AFB smear — negative.

Although there was a significantly greater incidence of h/o TB and
cavitation and lower incidence of increased interstitial markings/infil-
trates in TB pts, clinical/radiographic parameters are not clearly de-
fining in either group. Therefore, isolation of non-TB patients will oc-
cur for significant periods as a component of a TB control program.

56. Multivariate Analysis of Needlestick/Sharps Injuries (Sis) in 10
New York State (NYS) Hospitals, 1991. M. H, Mendelson,* L, Short, J.
Godbold, C. Schechter, X. Wu, B. R. Meyers, S. Z. Hirschman, L.
Chiarello. Mount Sinai School of Medicine, New York, NY and the New
York State Department of Health.

To target priorities for injury-reducing devices and strategies, an anal-
ysis of 1095 reported Sis from 10 NYS hospitals in 1991 was conducted.

58.7% Sis involved RNs, 16.4% MDs, 5.4% housekeeping/mainte-
nance, 3.5% laboratory workers. Hollow bore needles (N) accounted for
75.4% (N/syringe 32%, N/IV tubing (IVT) 15%, butterfly N 8.5%, un-
attached N 5.3%, vacutainer N 4.5%, IV stylet 3.9%); sutures 7.1%,
lancets 5.0%, scalpel blades 4.8% and glass 2.4%. Associated proce-
dures included: IV delivery related (IVDR) 18.3%, phlebotomy (PHD
11.7%, IM/SQ/ID injection 11.4%, finger/heelstick 5.7%, IV insertion
5.3%. Distribution of PHL Sis by sharp type: butterfly N 42.5%, vacu-
tainer N 30%, N/syringe 25.8%. 40.2% Sis occurred during the proce-
dure, 41% after use (5.8% recapping) and 17% during/after disposal.

Log-linear analysis showed the distribution of sharp types for MDs
(55% hypodermic (hypo) N, 25.8% butterfly N) to differ from that for
RNs (45% hypo N, 23% N/IVT) (p < .01). IVDR Sis occurred less fre-
quently in hospitals with vs. without safer systems: 1.47% vs. 12.53%
(hep lock); 4.48% vs. 8.04% (IVPB) (p < .05).

Devices with passive safety mechanisms should reduce Sis by 48-
67%, and active mechanisms by 17-28%. A significant impact on
worker safety and prevention of trsmsmission of blood-borne pathogens
should follow.

57. Antituberculous Agents Causing Acute Hepatic Failure: Success-
ful Treatment with Orthotopic Liver Transplantation (OLT). BR Mey-
ers, M Halpem, C Miller, M Schwartz, P Sheiner, MH Mendelson.
Divisions of Infectious Diseases & Transplantation, Mount Sinai Hos-
pital, N.Y.

Use of antituberculous agents for prophylaxis and treatment of sensi-
tive and resistant tuberculosis is increasing.

We report 7 patients with acute hepatic failure after receiving
tuberculosis prophylaxis (6) and treatment (1) in 1991 and 1992. There
were 2 males (1 Black, 1 Oriental) and 5 females (4 Hispanics, 1 Black),
mean age 41.2 years (5-69).

Possible cofactors for liver toxicity were found in 5:4 (medicines), 1
(ethanol), 1 (hepatitis B Virus), 1 (Epstein-Barr virus). Isoniazid (INH)
was given for 6-25.5 weeks (mean 12.8) (n = 6) and pyrazinamide
(PZA) for 3 weeks (n = 1) before presentation. Encephalopathy devel-
oped 1-57 days (mean 16.7) after stopping antituberculous agents.
Toxicity at presentation/and prior to OLT were: prothrombin time
25.5/34.1 (Sees), bilirubin 16.6/33.9 (mg/dL), AST 2101/397 (lU/L), ALT
2028/788 (lU/L). Pathology revealed massive hepatic necrosis. Patients

received OLT 11-61 days (mean 29.4) after presentation. Awaiting
OLT, 1 died of hemorrhage and infection. Survival: 4/5 OLT recipients
(mean 418.7 days, range 126-759 days), 3/4 patients with INH toxic-
ity; 1 patient with sequential INH and PZA for multiple drug resistant
tuberculosis prophylaxis survived following 3 OLTs (2 ABO incompat-
ible livers).

Conclusions. Since antituberculous-agent-induced acute hepatic
failure can develop rapidly once sjrmptoms occur, and survival after
OLT was 80%, referral for OLT is indicated early in management.
Other cofactors may enhance antituberculous agent toxicity and
should be looked for to identify patients at risk of acute hepatic failure.

58. Cytomegalovirus (CMV) Infections in Liver Transplant Patients
(pts): Decreased Incidence over 4 Years Associated with High Dose
Acyclovir and Intravenous Immunoglobulin (IVIG). BR Meyers, M
Halpem, C Miller, M Schwartz, P Sheiner, E Emor, MH Mendelson, L
Chodoff. Divisions of Infectious Diseiise, Liver Transplantation & De-
partment of Pharmacy, Mount Sinai Hospital, N.Y.*

From 9/88-8/90 (A), 98 pts underwent 118 orthotopic liver transplants
(OLT); from 9/90-11/92 (B) 301 pts underwent 347 OLT. Prophylaxis
for CMV in A was 200 mg oral acyclovir bid to all pts, and 500 mg/kg
IV IgG once weekly for 4 wks then qow x 4 to CMV negative recipients
(R-) who received CMV donor positive (D-I-) livers. During B, pts
received acyclovir 100 mg IV ql2h perioperatively then 800 mg po q6h
X 12 wks, plus rVIG to all R - pts. A and B were similar for: mean time
of surgery, underlying diseases, drug usage OKT-3 eind FK-506, and
age (47 yo in A, 43 in B).

In A/B, CMV caused viremia 1/4, hepatitis 19/15, pneumonia 3/3,
esophagitis 2/2, gastro/duodenal colitis 3/2 and retinitis 0/1. CMV dis-
ease developed more fi-equently in A 25/103 (24%) than B 22/289 (7.6%)
(p = 0.01), and in all sub-groups of A and B examined:

A(%)

B (%)

D-I-/R-I-

8/34 (23.5)

12/125 (9.6)

0.05

D + fR-

8/13 (61.5)

8/46 (17.4)

<0.001

D-fR +

7/33(21.2)

8/85 (9.4)

D-/R-

6/26 (23.1)

1/33 (2.9)

<0.025

There was a trend to develop CMV esirlier in A thsm B (51 vs 92 days;
p = 0.058). In both time periods (A -I- B), 35/276 (12.6%) of recipient
positive (R + ) and 23/119 (19.3%) of recipient negative (R- ) developed
CMV (p = NS). For all R-i- there was no association between donor
status and development of CMV; 20/159 (12.5%) D-i- and 15/119
(12.6%), D - (p = NS). For all R - there are more cases of CMV among
T>+ than D-; 16/59 (27%) D+ and 7/60 (11.6%) D- (p «.05). Rejec-
tions during A 86/124 (69%) and B 227/326 (70%) were similar (p =
NS). In A, CMV disease occurred in pts rejecting 17/64 (26%) and non-
rejecting (8/39) (20.5%) (p = NS). In B, CMV disease occurred in pts
rejecting 17/155 (11%) versus 5/134 (4%) without rejection (p « .05).

Conclusions. CMV infections were decreased in B versus A. CMV
prevention does not appear to reduce the incidence of rejection; with
rejection, CMV disease is less frequent when CMV prophylaxis is more
effective. CMV donor status was associated with increased infection
only in the negative recipient who received a positive organ. High dose
acyclovir w£is associated with less CMV disease in all recipient-positive
patients receiving donor positive liver. High dose acyclovir plus FVIG
was associated with less CMV disease in all recipient negative patients
regardless of CMV donor status.

59. Enterococcal Infections in Liver Transplant Patients: Increase in
Multiple Resistant Strains of E. faecium. BR Meyers, M Halpem, MH
Mendelson, C Miller, L Chodoff. The Divisions of Infectious Diseases,
Liver Transplantation & Pharmacy, Mount Sinai Hospital, N.Y.

From August 1988 to 1990 (A), 98 patients underwent 118 orthotopic
liver transplants and from September 1990 to 1992 (B), 301 patients
had 347 orthotopic liver transplants. Antibacterial prophyleixis for A
consisted of parenteral ampicillin '2 g) q6h or vancomycin (1 g) qd
(20%), and cefotaxime (1 g) q8h x 72 hours. Selective bowel deconttim-
ination began in B with oral gentamicin, nystatin and polymixin, given
qid, and as oral paste. Parenteral antibiotics in B were cefotaxime plus
vancomycin. The mean age, time of surgery, number of rejections and
underlying diseases were similar for A and B.

From August 1988 to September 1992 the rate of enterococcal in-
fections including E. faecalis and E. faecium for liver transplants var-
ied between 7.1 and 48%. The rate of all enterococcal infections/trans-

Voi. 60 No. 5

ABSTRACTS

429

plants were: 28% (A) and 26% (B), (p = NS). For A/B: 13%/14% of all
transplants (tx) had EFm isolated. Within B however there was in-
crease in E. faecium (E.Fm) isolates between Aug 90-Dec 91, 18/197 tx
(9.1%) and Jan-Dec 92, 39/157 tx (24.8%). During A/B: EFm
(12%)/(24%) isolates came from blood, (12%)/(9%) wound, (6%)/(7%)
bile, and (3%)/(9%) from peritoneum. During A/B, resistance of these
isolates were: (33%)/(84%) to ampicillin, (0%)/(49%) to vancomycin,
(33%)/(71%) to synergy screens with ampicillin plus streptomycin, and
(33%)/(59%) to ampicillin plus gentamicin. There were (0%)/(36%) that
were resistant to all above antibiotics.

Conclusions. The incidence of all enterococcal infections in A and
B were similar (p = NS). In B vs A: there were more E.Fm isolates
resistant to ampicillin (p = less than .01), vancomycin (p = less than
.001), and ampicillin/streptomycin (p = .01). Resistance to all antibi-
otics tested increased strikingly in 1992 with 79.5% of E.Fm. resistant
to all agents tested. Selective bowel decontamination may be associ-
ated with the isolation of multiple resistant strains of E.Fm. Both blood
isolates ofEFm and antibiotic resistance to these strains are increas-
ing; newer antimicrobial strategies will be necessary to treat these
serious infections. Doxycycline has proven efficacious in many of these
cases.

60. The Management of Persons Exposed to Multidrug-Resistant Tu-
berculosis; A Decision Analysis. David N. Rose, M.D. Divisions of Gen-
eral Medicine and Infectious Diseases and Department of Community
Medicine.

Purpose. To analyze various antituberculosis drug regimens to be used
as preventive therapy for persons exposed to multidrug-resistant
(MDR) tuberculosis (TB).

Methods. A decision analysis was performed for immunocompetent
and immunodeficient patients with 12 types of exposure to MDRTB
(various probabilities of infection with organisms of various sensitivi-
ties to antituberculosis drugs). Antituberculosis drugs were catego-
rized as first-line [isoniazid (INH), rifampin (RIF)], second-line (pyrazi-
namide, ethambutol, streptomycin, quinolones) and third-line drugs
(others); 13 drug combinations were studied. Second-line drugs were
assumed to be less effective and more toxic than INH, and third-line
drugs more so. A Markov model was used to compare the number of
deaths over 10 years among 30-yeaT old people with and without each
regimen. 1400 combinations of patient type, exposure type, regimen,
and assumptions [favorable, skeptical, and mid-range (base case)] were
studied.

Results. Recent tuberculin converters will have greatly reduced
mortality from preventive therapy under most circumstances. For ex-
ample, immunocompetent persons with 0.5 probability of infection
with drug-sensitive TB and 0.5 probability with MDRTB (resistant to
INH and RIF but sensitive to second-line drugs) have 50% less mor-
tality with a regimen of INH and a second-line drug (than without
preventive therapy) and 44% less mortality with a second-line drug
alone, under base case assumptions. These regimens are effective even
under a combination of the most skeptical assumptions regarding drug
effectiveness and toxicity. In certain circumstances, such as exposures
to organisms resistant to second-line drugs, and under the most skep-
tical Eissumptions, the optimal strategy is no preventive therapy. Im-
munodeficient persons have even more benefit from various regimens,
and fewer circumstances when no preventive therapy is best. This is
true even if these persons are anergic, and the probability of recent
infection is only 0.5. When there is even a low probability of exposure
to INH-sensitive organisms (0.25 probability or higher), the best strat-
egy is preventive therapy with INH plus second-line or third-line
drugs.

Conclusions. Preventive therapy benefits most persons exposed to
MDRTB. The optimal strategy depends upon best estimates of the drug
sensitivity of the organism(s). Even under very skeptical assumptions
regarding preventive therapy, certain regimens are beneficial for
many exposures.

61. The Tuberculin Skin Test and Isoniazid Preventive Therapy. David
N. Rose, M.D., Clyde B. Schechter, M.D., Jack J. Adler, M.D. Divisions
of General Medicine, Infectious Diseases and Pulmonary Medicine and
Critical Care, Department of Medicine and the Department of Com-
munity Medicine.

Background. The tuberculin skin test is the only diagnostic test for
latent Mycobacterium tuberculosis infection. Latent infection can acti-
vate at any time, but isoniazid preventive therapy can prevent activa-
tion. The tuberculin test's inability to accurately identify M. tubercu-

losis infection and isonieizid hepatotoxicity complicate the decision to
prescribe isoniazid.

Purpose. To examine the interpretation of the tuberculin skin test
by addressing two questions: (1) what is the probability of Af. tubercu-
losis infection at each tuberculin reaction size, and (2) for which tuber-
culin reactors do the benefits of isoniazid outweigh the risks?

Design. We modeled the probability of infection as a function of
tuberculin reactivity size, the prevalence of nontuberculous mycobac-
terial infection, and the presence or absence of risk factors for recent or
remote infection. The preventive therapy decision depends on these
and other variables. We emalyzed the same studies referenced by the
Centers for Disease Control's advisory committees in their recommen-
dations. We also surveyed advisory committee members about uncer-
tain variables and then conducted a decision analysis.

Results. The probability of M. tuberculosis infection varies from
0.01 for people from southern states with 2 to 4 mm tuberculin reac-
tions to 1.0 for all persons with reactions of 15 mm or more. The ben-
efits of 24 weeks of isoniazid preventive therapy outweigh the risks for:
all persons 5 to 65 years old with s=10 mm reactions, except some
persons with nonhematologic neoplasms; most persons with 5 to 9 mm
reactions who have clinical or social factors that raise the tuberculosis
risk and some with no additional risk factor; and some persons with 2
to 4 mm reactions, especially those with HIV infection, fibrotic lesions
on chest x-ray, or recent close contact with active cases. The benefits
are greater with 52 weeks of preventive therapy.

Conclusions. Tuberculin reactivity only indicates a probability of
M. tuberculosis infection. Current standards for labeling reactions as
"positive" misclassify many reactors, and these classifications should
be avoided. Twenty-four or 52 weeks of isoniazid preventive therapy
would benefit many people not covered by current recommendations.

62. Tuberculosis Sensitivity Patterns, Predictors of Multidrug Resis-
tance, and Implications for Initial Therapeutic Regimens at a New
York City Hospital. David N. Rose, Andr6 C. Weltman. Divisions of
General Medicine and Infectious Diseases and Department of Commu-
nity Medicine.

Objective. To identify predictors of multidrug resistant (MDR) tuber-
culosis (TB); and to analyze the potential effectiveness of the standard
initial 4-drug regimen and other regimens.

Design. A case-control study of all tuberculosis patients at Mount
Sinai Hospital in 1991 and 1992, and a descriptive analysis of the
isolates' susceptibility patterns to determine potential effectiveness of
initial drug regimens. A potentially effective regimen is assumed to
have at least two drugs active against an isolate. MDR-TB is defined as
resistance to isoniazid (INH) and rifampin (RIF).

Results. Isolates from 172 patients were studied; 28.5% were re-
sistant to INH, 21.1% to RIF, 15.9% to ethambutol (EMB), 11.3% to
pyrazinamide (PZA), 18.6% to streptomycin, 13.1% to ethionamide
(ETH), 2.4% to cycloserine (CS), and none to capreomycin (CAP) or
ciprofloxacin (CIP). Thirty-two (18.6%) isolates were resistant to both
INH and RIF. AIDS, HIV seropositivity, female gender, residence in
the Bronx, and race were significantly associated with MDR-TB. The
4-drug regimen of INH, RIF, PZA, and EMB is potentially effective for
81% to 85% of all patients. The subgroup of patients with the most
benefit from this combination is patients without AIDS, from 88% to
89% potential effectiveness. The subgroup with the least benefit is
patients with AIDS and prior TB therapy, from 60% to 80% potential
effectiveness. The substitution of ETH, CIP, CAP, or CS for EMB im-
proves potential effectiveness moderately. Most 5-drug regimens do not
have substantially higher potential effectiveness than 4-drug regi-
mens. Only 5- or 6-drug regimens containing two or three of CAP, CIP,
CS, and ETH offer substantially higher potential effectiveness.

Conclusions. We found high levels of MDR among the TB patients
at our hospital. MDR-TB in these patients is associated with HIV in-
fection, AIDS, female gender, residence in the Bronx, £ind certain racial
groups. The 4-drug initial regimen used in New York City is inade-
quate for our patients. Only certain 5- or 6-drug regimens have the
potential for substantial effectiveness as initial therapy.

63. Screening for HIV Risk Factors and Referral for HIV Testing in
the Medical Primary Care Clinic at Elmhurst Hospital Center, 1993.

A. D. Urena, M.D.

Primary care physicians are in a unique position to screen and educate
patients at risk for HIV infection. A study was conducted to evaluate
screening and referral practices of physicians in the Medical Primary
Care Clinic (MPCC). MPCC is a public based clinic providing compre-

430

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

hensive primary care services with 28,000 visits a year by 9,000 pa-
tients.

A random sample of 73 charts was selected. In 40 charts (55*^) the
standard history form included questions on HIV risk factors; of these,
39 charts l98'J) contained documentation of risk factor historical in-
formation; one (d'ye) did not. Five patients (13%) were identified as
being at risk for HIV infection. Three (60%) had been recently tested;
of the two not tested (40*^), documentation of referral for testing was
noted in the chart for one patient (SO*^) and no documentation of dis-
cussion or referral was found for the other patient (50%). In 33 (45%) of
the total charts reviewed, the initial history was taken prior to 1990
when the standard history form in use did not include printed ques-
tions on HIV risk factors. None of these charts contained initial or
updated historical information concerning HIV risk factors.

Patients initially evaluated prior to 1990 should have HIV risk
factor screening, all patients may benefit from interval updating of risk
history. While increased physician and patient awareness since 1990
may have resulted in the 98% rate of screening for HIV risk factors,
these results suggest that use of the amended history form may influ-
ence physician behavior.

Further study is suggested to evaluate the effectiveness of screen-
ing and referral in the MPCC. This is an important clinical issue that
is relevant to all physicians especially those working in the primary
care setting.

64. Caesarean Section to Reduce Perinatal Transmission of Human
Immunodeficiency Virus: A Meta-Analysis. Paolo Villari, MD, Cathie
Spino, MD, Thomas C. Chalmers, MD, Joseph Lau, MD, Henry S.
Sacks, Ph.D., MD. Harvard School of Public Health, Boston; Facolta' di
Medicina e Chirurgia, Universita' degli Studi "G. D'Annuzio," Chieti;
New England Medical Center, Boston; Mount Sinai School of Medicine,
New York.

Objective. Individual epidemiologic investigations into the association
between type of delivery and perinatal HIV transmission have been
suggestive but inconclusive. Meta-analysis was used in an attempt to
establish if there is, at present, adequate evidence concerning the ef-
fectiveness of caesEirean section in reducing vertical HIV transmission
rates.

Methods. The MEDLINE data retrieval system and other sources
were used to identify studies containing data on the relationship be-
tween type of delivery and vertical HIV transmission. No randomized
control trials were located. Seven of the 11 cohort studies identified
were included in the meta-analysis. Crude or, when available, adjusted
data were extracted and pooled.

Results. The overall weighted risk of perinatal HIV infection was
21.8% after vaginal and 15.8% after caesarean delivery. This difference
was statistically significant. Pooling data of all studies yielded an over-
all odds ratio of 0.66 (95% CI: 0.48, 0.91) (Random effects model; Der-
Simonian and Laird method). Approximately 16 (95% CI: 43, 10) HIV-
infected women must deliver by caesarean in order to prevent 1 case of
HIV perinatal infection.

Conclusions. Results of this study show that performing elective
caesarean section in HIV-infected women is potentially a very effective
procedure. However, the non-experimental nature of the individual
studies leads us to conclude that randomized control trials are indi-
cated before setting specific guidelines for mode of delivery in HIV-
infected women.

Geriatrics

65. 25-Hydroxyvitamin D in Elderly Nursing Home and Enriched
Housing Residents in New York City. C. Feiner, M. Mulvihill, B. Tay-
lor, W. Bauman, H. Smith, E. Weintraub, P. Tsitouras. Dept. of Geri-
atrics, The Mount Sinai School of Medicine, New York, NY, Jewish
Home and Hospital for Aged, Bronx, NY, and Bronx VA Medical Cen-
ter, Bronx, NY.

Although serum 25-hydroxyvitamin D (25-OH) has not been demon-
strated to be low in American community dwelling elderly, there is
conflicting data concerning levels in institutionalized elderly.

We examined serum 25-OH in 16 female Caucasian nursing home
residents exposed to small amounts of sunlight («15 min/day for pre-
ceding three months) vs. 16 age-matched enriched housing residents
exposed to large amounts (ss60 min/day). History of sunlight exposure.

diagnoses, medications and functional status were obtained by chart
review and interviews. Dietary data were obtained by 3 day videotaped
assessment. Fasting blood assays were done in September and October;
25-OH w£is measured by RIA.

Mean 25-OH level was 20.0 ±7.4 ng/ml in the high exposure Eind
18.4 ± 6.5 ng/ml in the low exposure group (p = 0.35, NS). Low 25-OH
levels (sl9 ng/ml) were present in 31% of residents with high sunlight
exposure and in 56% of residents with low sunlight exposure. Mean vit.
D intake was 252 ± 176 lU in the high exposure and 305 ± 226 Hi in
the low exposure group (P = 0.52, NS). There was a significant corre-
lation (r = 0.45, p = 0.016) between 25-OH £ind vit. D intake when
both sunlight exposure groups were combined. There was no signifi-
cant correlation between serum calcium (r = 0.13, p = 0.48) and serum
phosphorus (r = 0.20, p = 0.27) versus 25-OH.

We conclude that there is a high prevalence of low 25-OH levels in
nursing home and enriched housing residents which, as opposed to
sunlight exposure, seems to be a function of vitamin D intake. No
metabolic consequence of low 25-OH was detected but data suggest a
risk for hypovitaminosis D. Closer attention to vitamin D intake in
these populations is indicated.

66. Drug Treatment of Hypertension in the Elderly: A Meta-Analysis.

Jorge T. Insua, MD, Henry S. Sacks, PhD, MD, Tai-Shing Lau, PhD,
Joseph Lau, MD, Dinah Reitman, MPS, Thomas C. Chalmers, MD.
From the Departments of Community Medicine (J. I., H.S.), Geriatrics
and Adult Development (J.I.), Medicine, Biomathematical Sciences
(H.S., T-S.L, D.R.), Mount Sinai School of Medicine of CUNY, New
York, NY: the Technology Assessment Group, Harvard School of Pub-
lic Health, Boston, MA (T.C.C.), and the Boston Department of Veter-
ans Affairs (J.L.).

Purpose. To examine by meta-analysis the effect of antihypertensive
drug treatment on total mortality, cardiovascular mortality, and cause
specific morbidity and to analyze the effect of severity of illness and age
on outcome, in the elderly population.

Data sources. A literature search for randomized controlled trials
of hypertension in the elderly.

Study selection. Randomized control trials of drug treatment of
hypertension with endpoints of all cause and/or cardiovascular mortal-
ity reported separately for elderly patients.

Data extraction. Mortality and/or morbidity endpoints in patients
over 59, were pooled by determination of typical odds ratio. A meta-
regression was used to study heterogeneity.

Results. Nine major trials were identified, with a total of 15,559
patients over 59. The ranking of the trials by severity of illness was
related to treatment outcome. Death rates varied between 2.7% and
77.2%; stroke and coronary mortality increased with severity of illness
rank (p < 0.0001). Meta-analysis showed that overall, treated patients
had approximately 12% reduction in mortality, (odds ratio (OR) 0.88,
95% confidence interval (CI) 0.80-0.97, 953 vs 1069 events, p =
0.0092). There was 22% reduction in vascular deaths (OR 0.78, 95% CI
0.69-0.89, 462 vs 584, p = 0.00017), 36% for stroke (OR 0.64, 95% CI
0.49-0.82, 94 vs 149, p = 0.0005), and 25% reduction of coronary heart
disease mortality (OR 0.75, 95% CI 0.64- 88, 263 vs 350, p = 0.00055).
Coronary morbidity was reduced 15% (OR 0.85, 95% CI 0.73-0.99, 325
vs 379, p = 0.036) and stroke morbidity reduced 35% (OR 0.65, 95% CI
0.55-0.76, 247 vs 382, p < 0.00001).

Conclusion. Overall, treatment of hypertension in elderly patients
produces a significant benefit in total mortality and cardiovascular
morbidity and mortality. However, this benefit may be reduced in the
more severely ill elderly.

67. Validation of the GDS (YESAVAGE) Depression Screen in the
Nursing Home. Shawn McGivney MD, Brian Taylor PhD, and Michael
Mulvihill DPH. Jewish Home and Hospital for Aged, New York, and
Mount Sinai Medical Center, New York, NY.

The GDS (Geriatric Depression Scale) has demonstrated validity
among ambulatory elderly but is less useful in nursing home (NH)
populations, probably because of high rates of cognitive impairment.
We therefore sought the lowest level of Mini Mental Status Exam
(MMSE) score for which the GDS would remain valid.

A total of 66 of 168 newly admitted residents to the NH were able
to complete psychiatric assessment, undergo an MMSE and GDS. The
psychiatrist and testers (who were virtually all non-MDs) were blinded
to each others' results. Using a cut off of 10 or greater on the GDS to
indicate depression, the validity of the GDS with the psychiatric diag-
nosis was made at ever decreasing levels of cognitive function.

In all residents the sensitivity and specificity were 63% and 83%

Vol. 60 No. 5

ABSTRACTS

431

respectively. When only those with an MMSE of greater than or equal
to 15 were included, the corresponding values were 84% and 91%. The
proportion of the NH residents screened meeting this cut-point was

64%.

This two step procedure of first selecting those with MMSE greater
than or equal to 15 increases the utility of the GDS in NH's and should
improve the diagnostic process for the underdetected problem.

68. Behavioral and Cognitive Changes in Vasopressin-Secreting
Transgenic Mice. Myron Miller, Vahram Haroutunian,* Michelle
Wiltshire-Clement,* and Shigeki Kawabata.* Depts. of Geriatrics &
Adult Devel., & Psychiatry, The Mount Sinai Med. Ctr., NY, NY.

Vasopressin effects on the central nervous system of rats and mice
include enhancement of emotionality, attention and arousal. We have
developed a line of transgenic mice which overexpresses the rat gene
for vasopressin with resultant increases in vasopressin content in brain
and the peripheral circulation.

Locomotor behavior in 8 to 1 1 months-old male mice homozygous
and heterozygous for the rat transgene, as well as normal controls, was
evaluated in an open field apparatus for 5 consecutive days and again
9 and 10 days later.

Total distance traveled by the animals during days 1 to 4 declined
each day for each group, indicating habituation. The distance travelled
on day 4 was significantly lower in homozygous mice (3912 ± 991 cm,
SEM) than in heterozygous (6773 ± 552 cm, p < 0.005) or normal
animals (6859 ± 853 cm, p < 0.02). Introduction of a novel object into
the center of the field on day 5 resulted in increased center field ex-
ploration which was most pronounced in the heterozygous group.
Retesting on day 14 demonstrated retention of long-term habituation
by all 3 groups, with the homozygous mice showing the greatest effect.
Learning and memory capacity was assessed by passive avoidance test-
ing using a light-dark shuttle box. Initial latencies did not differ be-
tween the 3 groups. Retesting 72 hrs after foot-shock showed no sig-
nificant differences among the mice on retention test performance,
indicating equal learning and memory for the passive avoidance task.
Following an 18-hour period of fluid deprivation, homozygous mice
drank significantly more 10% sucrose solution than did control ani-
mals, indicative of diminished neophobia.

Thus, mice with increased vasopressin synthesis exhibited altered
behavior consistent with enhanced attention and alertness but without
change in learning and memory ability. These observations support
prior data on vasopressin effects on cognitive function and suggest that
the transgenic mice are a useful model for investigation of hormonal
actions on the central nervous system.

were corrected for plasma total or LDL cholesterol. Plasma retinol,
a-tocopherol and other carotenoids (such as lycopene) did not differ
significantly.

Conclusion. Plasma p-carotene is relatively increased by heavy
alcohol consumption, whereas liver damage, especially cirrhosis, has
the opposite effect. Malabsorption secondary to pancreatic insufficiency
is apparently not responsible, nor can one incriminate lipid malabsorp-
tion, since other liposoluble compounds were not affected similarly. In
these patients, p-carotene supplementation may be justified, but this
should be coupled with control of drinking because of possible hepato-
toxic alcohol/p-carotene interactions.

70. 3T3 Fibroblasts Transfected with an Aspartate Aminotransferase
cDNA Express Both Plasma Membrane Fatty Acid Binding Protein
and Saturable Fatty Acid Uptake. LM Isola, S-L Zhou, C-L Kiang, DD
Stump, and PD Berk. Depts. of Medicine & Biochemistry, Mt. Sinai
School of Medicine, New York, NY 10029.

Uptake of long chain free fatty acids (FFA) exhibits kinetic features of
facilitated transport in hepatocytes and certain other cells but not in
fibroblasts. Several lines of evidence (Ann Rev Nutr 9:252) suggest
that a 43 kDa plasma membrane fatty acid binding protein, FABPpm,
is the FFA transporter. FABPpm, surprisingly, was found to be iden-
tical to the mitochondrial isoform of aspartate aminotransferase
(mAspAT) (PNAS 87:3484; Stump DD, this meeting). To document its
role in FFA transport, we constructed plasmid pMKAAT2, a modified
pGEM3-Z containing a full length mAspAT cDNA downstream of a
Zn * * -inducible metallothionine promoter.

3T3 fibroblasts were co-transfected with pMKAAT2 and pFR400, a
plasmid conveying methotrexate (MTX) resistance, by Ca3(P04)2-DNA
co-precipitation. Transfectants were selected in MTX, cloned, and ex-
posed to increasing MTX concentrations to induce gene amplification.
Stably transfected clones were characterized by Southern blotting;
those with highest copy numbers of pFR400 alone (pFR400) or pFR400
plus pMKAAT2 {pFR400IAAT) were expanded for use in later studies.
Cell surface expression of FABPpm was examined by immunofluores-
cence using rabbit anti rat liver FABPpm.

Neither untransfected 3T3 nor pFR400 cells expressed surface
FABPpm, though both exhibited punctate intracellular fluorescence,
presumably in mitochondria. By contrast, plasma membrane immuno-
fluorescence was clearly observed in pFR400IAAT cells, particularly
after culture in 100 jjlM Zn * * . [^Hl-oleate uptake was studied by rapid
filtration from media containing 500 jjlM BSA etnd 50-1000 p-M total
oleate (unbound oleate:6. 5-430 nM). Initial velocities were determined
over 45 sec, and the data fitted with the SAAM program to the sum of
a linear plus a saturable function. Results (Table) indicate a 2-fold
increase in Vmax (fmol/sec/50,000 cells) for oleate

Liver Diseases

69. Plasma B-Carotene Correlates Positively with Alcohol Intake, but
Is Decreased by Alcoholic Liver Disease. S. Ahmed, M. A. Leo, C. S.
Lieber. Alcohol Research and Treatment Center and Section of Liver
Disease and Nutrition, Bronx VA Medical Center and Mt. Sinai School
of Medicine, New York, NY.

Alcoholics have low fasting plasma p-carotene levels, probably due to
poor dietary intake, but the possibility that alcohol might also have a
more direct effect remained to be assessed.

Therefore, plasma carotenoids were determined by HPLC and
their levels were compared with the average alcohol consumption over
the preceding 3 months.

Whereas alcoholics had generally lower plasma p-carotene concen-
trations than controls, the heavy drinkers (>200 g/day) had about
twice the p-carotene levels than those drinking less (p < 0.01). There
was a significant correlation between plasma p-carotene and alcohol
intake (r = 0.6, p < 0.001), even when the p-carotene values were
expressed per plsisma cholesterol, to correct for possible lipoprotein
changes (r = 0.67, p < 0.001). Thus heavy alcohol consumption ap-
peared to increase plasma p-carotene levels but, since it is associated
with liver damage, the latter could also have played a role. To assess
the influence of liver disease, p-carotene beadlets (30-60 mg per day)
were given for three days to hospitalized alcoholics fed controlled diets.
Patients with cirrhosis had a much lower plasma p-carotene response
than those without; the latter in turn responded with lower p-carotene
levels than controls (drinking <5 g ethanol/day). These differences
were not eliminated by pancrease supplementation or when the values

uptake in pFR400

pFR400/AAT

-Zn

Vmax
Km

152 ± 24
113 ± 29

+ Zn

141 ± 15
122 ± 22

-Zn

+ Zn

286 ± 97
106 ± 57

809 ± 193
118 ± 50

pFR400IAAT cells compared to pFR400, with a further 2.8-fold in-
crease in presence of Zn * * . Zn * * had no effect in pFR400 controls (p
> 0.5). The 5.7-fold increase in Vmax between pFR400 and
pFR400IAAT in presence of Zn* * was highly significant (p < 0.025).
Km's in pFR400IAAT and pFR400 cells with/without Zn* * did not
differ.

Conclusions. mAspAT and FABPpm are identical, and mediate
saturable FFA uptake.

71. Serum Tissue Inhibitor of Metalloproteinase (TIMP) Is Increased
in Precirrhotic and Cirrhotic Alcoholics and Can Serve as a Marker of
Fibrosis. J. Li, A. Rosman, M. Leo, Y. Nagai,* C. Lieber. Alcohol Re-
search and Treatment Center, Bronx, VA Medical Center and Moimt
Sinai School of Medicine, Bronx, New York 10468. *Fuji Chemical
Industries Ltd., Toyama, Japan.

Hepatic fibrosis reflects an imbalEince between collagen production and
degradation. One of the contributory factors to the development of he-
patic cirrhosis is a decrease in collagenase activity (Maruyama et al.,
BBA 1981;658:121).

To determine whether this may be related to inhibition of tissue
collagenase, TIMP was measured with a sandwich ELISA assay
(Kodama, Matrix 1989;9:1) in the serum of 16 healthy controls and 42

432

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

alcoholic patients with biopsy proven liver disease, namely steatosis
without fibrosis (N = 13), perivenular fibrosis (PVF, N = 12), and
septal fibrosis and/or cirrhosis (N = 17). Liver biopsies were evaluated
for fibrosis by a pathologist blinded to the clinical and laboratory fea-
tures.

Serum TIMP values strongly correlated with fibrosis (r, = 0.74, P

< 0.001) and only modestly with inflammation scores (r^ = 0.38, P <
0.01). Compared to controls (mean value ± SEM, 177 ± 12 ng/ml),
serum TIMP was significantly elevated in PVF (348 ± 27, p < 0.005)
and in septal fibrosis and/or cirrhosis (458 ± 52, P < 0.001) but not in
steatosis (204 ± 17, P > 0.5). Thus, the rise in TIMP does not appear to
be a mere consequence of the cirrhosis, since it was present already at
a precirrhotic stage. We also compared the diagnostic utility of serum
TIMP to an RIA method of serum procoUagen-III-peptide (PIIIP) mea-
surement (Behring, Germany) another proposed marker of fibrosis. In
contrast to serum TIMP, serum PIIIP was significantly elevated only in
the septal fibrosis and/or cirrhosis group (1.41 ± 0.24 U/ml, P < 0.01)
but not in the PVF (0.71 ± 0.10, P > 0.5) when compared to healthy
controls (0.66 ± 0.03). It also was not increased in the steatosis group
(0.56 ± 0.06). By defining the threshold as the upper value of the
steatosis group (resulting in a specificity of 100%), serum TIMP was
elevated in 50% of patients with PVF and 88% of patients with exten-
sive fibrosis (septal fibrosis and/or cirrhosis). The overall sensitivity of
serum TIMP for detecting either PVF or extensive fibrosis was signif-
icantly greater than that of serum PIIIP (72% vs. 41%, P < 0.05). At a
specificity of 85% (i.e., 85% of alcoholics with steatosis have values
below the selected threshold), the overall sensitivity for detecting ei-
ther PVF or extensive fibrosis was 97% for TIMP vs. 72% for PIIIP (P

< 0.05). Receiver operating characteristic (ROC) analysis was used to
assess the abilities of the two tests to discriminate PVF or extensive
fibrosis fi-om steatosis. The area under the ROC curve for TIMP was
significantly greater than for PIIIP (0.96 ± 0.03 vs. 0.79 ± 0.07, P <
0.03).

Conclusions. 1) serum TIMP is increased in alcoholic cirrhosis and
may play a role in its pathogenesis through inhibition of collagenase
activity. 2) In alcoholics, serum TIMP can serve as a marker of precir-
rhotic and cirrhotic states. 3) This test is more sensitive in detecting
either PVF or extensive fibrosis and offers better discrimination from
steatosis than serum PIIIP.

72. Phosphatidylcholine Corrects the Ethanol-Induced Decrease in
Hepatic Phosphatidylethanolamine Methyltransferase (PEMT) Activ-
ity and Protects Against Cirrhosis. C. S. Lieber, S. Robins,* M. A. Leo.
Alcohol Research and Treatment Center, Section of Liver Disease and
Nutrition, VA Medical Ctr, Bronx, NY, Mt. Sinai School of Medicine,
New York, NY and *VA Medical Ctr, Boston, MA.

Ethanol abuse results in striking membrane alterations which may
play a key role in the pathogenesis of cirrhosis. Membranes consist
mainly of phospholipids, primarily phosphatidylcholine. The synthesis
of these lipids depends, in part, on PEMT, reported to be decreased in
patients with cirrhosis. It is not known, however, whether this decline
is a consequence of the cirrhosis or precedes it, and to what extent it is
eissociated with phospholipid alterations. We also wondered whether
dilinoleoyl phosphatidylcholine (which is highly bioavailable) can cor-
rect the enzyme defect as well as the ethanol-induced phospholipid
abnormalities, and whether this could result in protection against cir-
rhosis.

This was studied in a baboon model of alcohol-induced cirrhosis.
PEMT was measured in sequential percutaneous needle liver biopsies
by the conversion of phosphatidylethanolamine to phosphatidylcho-
line, using radioactive S-adenosylmethionine.

Chronic alcohol consumption (1-6.5 yrs) resulted in the develop-
ment of septal fibrosis in 10/12 baboons (with cirrhosis in 2); it signif-
icantly decreased hepatic phospholipids, phosphatidylcholine and
PEMT already at non-fibrotic stages. These effects were fully pre-
vented in 8 baboons supplemented with 2.8g/1000 kcal of a preparation
rich in dilinoleoyl phosphatidylcholine. There were significant (p <
0.001) correlations between PEMT and either hepatic phosphatidylcho-
line (r = 0.678) or total phospholipids (r = 0.662).

Conclusions. 1) Alcohol consumption diminishes PEMT prior to
the development of cirrhosis and decreases hepatic total phospholipids
and phosphatidylcholine, key components of cell membranes; this may
promote hepatocyte injury and thus trigger fibrosis. 2) Administration
of dilinoleoyl phosphatidylcholine restores PEMT and corrects phos-
pholipid and phosphatidylcholine depletions, thereby possibly contrib-
uting to the protection afforded by dilinoleoyl phosphatidylcholine
against alcoholic liver injury and the ensuing cirrhosis. This striking

protective effect of dilinoleoyl phosphatidylcholine against alcoholic
cirrhosis in the baboon is now being verified in a multicenter, random-
ized, double-blind trial in man.

73. Alcohol Consumption Induces Fatty Acid a>-Oxidation, with a
Greater Increase in Males than in Females. X. Ma, E. Baraona and
C. S. Lieber. Alcohol Research & Treatment Center, Mount Sinai
School of Medicine & Bronx VAMC, New York, NY.

Women are more vulnerable than men to develop alcoholic liver injury.
Chronic ethanol administration to rats produced striking accumulation
of non-esterified fatty acids in the liver of female, but not male, rats.
This was associated with an ethanol-induced increase in cytosolic fatty
acid binding protein (L-FABPc) and microsomal esterification much
smaller in females than in males, despite similar inhibition of mito-
chondrial p-oxidation. Since products of lo-fatty acid oxidation have
been found to increase both L-FABPc and peroxisomal ^-oxidation, and
ethanol induces microsomal activities, we wondered whether alcohol
consumption induces cytochrome P450 4Al-mediated u)-oxidation and
whether this response depends on gender.

To study this, 12 male and 12 female littermates of the same age
were pair-fed liquid diets containing 36% of energy either as ethanol or
as additional carbohydrate for 4 weeks. The initial step of oi-oxidation
was assessed by the microsomal formation of 12-hydroxydodecanoic
acid (measured by GC/MS) from lauric acid.

In the pair-fed controls, the rate of (o-hydroxylation was higher in
females than in males (p < .05). However, ethanol feeding markedly
increased the rate of o)-hydroxylation in males (248 ±21 |ig/30 min/g
liver vs 131 ± 12 in controls; p < .01), but not in females (209 ± 16 vs
168 ± 19; N.S.). Since this process is catalized by cytochrome P450 4A1,
we assessed the effects of ethanol feeding and gender on this cy-
tochrome, using a sheep polyclonal IgG (kindly provided by Dr. G.
Gibson), which specifically recognizes the P450 4A1. Scanning densi-
tometry of Western blots from microsomal proteins revealed a 46%
increase in P450rVAl in ethanol-fed males (p < .02) with only 19%
increase in the females. Despite competition of ethanol and oj-hydroxy
fatty acids for alcohol dehydrogenase (ADH), the dicarboxylic acid
products of this reaction accumulated in the liver of alcohol-fed male,
but not female rats, probably due to the much higher affinity of ADH
for o)-hydroxy fatty acids than for ethanol. These finding paralleled
those on L-FABPc and fatty acid esterification, previously reported in
similar treated rats.

In conclusion, alcohol consumption induces microsomal P450 4A1
and increases fatty acid (o-oxidation. By serving as an alternate oxida-
tion pathway and by increasing fatty acid binding and esterification,
the increase in (u-oxidation may compensate, at least part, for the def-
icit in fatty acid oxidation due to the ethanol-induced injury of the
mitochondria, but less efficiently in females than in males, leading to
potentially deleterious accumulation of free fatty acids in the female.

74. Recurrence of Hepatitis B Viral (HBV) Disease in Liver Trans-
plant Recipients: Impact of Pre-Transplant HBV DNA Screening by
Liquid Phase Hybridization Assay. A Min, A Borcich, L Leung, P
Hytiroglou, S Emre, C Miller, S Thung, HC Bodenheimer. Depts. of
Medicine, Pathology and Surgery, Mt. Sinai Medical Center, New
York, NY.

Previous studies suggest almost universal recurrence of HBV disease
following liver transplantation (OLT) in patients with active viral rep-
lication. However, HBV recurrence in patients who are serum HBV
DNA negative pre-OLT has been less predictable. To fiirther charac-
terize HBV re-infection in such patients, we reviewed our experience
over the period 1988-93 at Mount Sinai.

Methods. Of the 22 patients transplanted for chronic HBV disease,
serum HBV DNA level post-OLT and/or follow-up (FAJ) liver biopsy
specimens were available on 19 patients (14M and 5F) surviving >1
month (mo). The mean duration of F/U for these 19 patients was 14.3
mo (range 1.3-39). The status of HBeAg, HBeAb and HBV DNA of the
recipients was reviewed. Serum HBV DNA was measured using a liq-
uid phase hybridization assay with negative cutoff at <1 pg/ml. No
patient received immunoprophylaxis with IV HBIG.

Results. Mean age at time of OLT was 43.6 yrs (range 24-60).
HBeAg was ( - ) in 17/18 patients (94%) tested. HBV DNA was ( - ) in
16/18 patients (89%). 15/19 patients are alive at a F/U of 15.5 mo (range
4.5-39). Histologic recurrence occurred in 11/19 patients (58%) at a
mean of 6.0 mo post-OLT, but 7 patients (6 HBV DNA ( - )) have no
recurrence with a F/U of 16.6 mo (range 4.5-39). Four patients died (2
due to HBV) at a F/U of 9.8 mo (range 1.3-17). Of 16 patients who were
HBV DNA ( - ) pre-OLT, HBV DNA remained undetectable in 8 pa-

Vol. 60 No. 5

ABSTRACTS

433

tients (50%) up to 26 mo post-OLT; histologic recurrence occurred in
9/16 patients (56%) and 13/16 (81%) are alive at 15.2 mo F/U (range
4.5-39). Both patients who were HBV DNA ( + ) pre-OLT had histo-
logic recurrence; 1 died of suicide and one is alive at 17 mo. Retrans-
plants were done for primary nonfunction (4) and for hepatic artery
thrombosis (1).

Conclusions. The absence of serum HBV DNA (<1 pg/ml) is asso-
ciated with favorable early survival, and although disease recurrence
is common, many patients (44%) remain free of HBV disease. The high
rate of recurrent hepatitis indicates the need for effective anti-viral
therapy post-OLT.

75. Why Does a More Concentrated Alcoholic Beverage Yield Lower
Blood Alcohol Levels in the Fed State? Rajesh Sharma, M.D., R. Thom-
as Gentry, Ph.D., Zev Chayes, M.D., and Charles S. Lieber, M.D. Dept
of Medicine, Mt Sinai School of Medicine and the Alcohol Research and
Treatment Center, and Nuclear Med Service, Bronx VAMC, New
York, NY.

Contrary to general expectations, more concentrated alcoholic bever-
ages produce lower blood alcohol concentrations (BAC) than dilute bev-
erages when both are consumed in the fed state (Roine et al.. Alcohol-
ism: Clin Exp Res 17:709-711, 1993).

To assess the role of gastric emptjdng, five healthy men were given
alcohol (0.3 g/Kg, consumed as a 4% and 10% w/v solution in orange
juice) with ®^'"Tc-DTPA one hour after a standard meal.

The 10% alcohol significantly slowed gastric emptying (half-time
73 ± 10 min) compared to 4% alcohol (37 ± 4 min, p < 0.01). The
absence of a significant effect using the same volumes without alcohol
(half-time of 40 ± 5 min with 235 mL us 35 ± 3 min with 588 mL, n.s.)
demonstrated that the effect of concentration was a consequence of the
alcohol and not the difference volume. To assess whether there was also
an effect of alcohol concentration on the First Pass Metabolism (FPM)
of alcohol (which occurs primarily in the stomach), eight subjects (in-
cluding the 5 above) were given, on different days, the same dose of
alcohol by intravenous infusion, and by oral consumption (4 and 10%
w/v). The quantity of alcohol absorbed by the oral route was calculated
by integration of the blood alcohol curves, using the in vivo elimination
parameters obtained from the intravenous curve for each subject. This
procedure, which is independent of rate of absorption (Gentry et al..
Gastroenterology 92:A11, 1992), indicated that FPM (the difference
between the quantity absorbed and the dose administered) increased
from 28 ± 23 mg/Kg with 4% alcohol to 108 ± 12 mg/Kg with 10% (p
< 0.01). This effect probably results ft-om enhanced gastric ADH ac-
tivity at higher ethanol concentrations as well as the prolonged expo-
sure of alcohol to the gastric mucosa.

In conclusion, when consumed after a meal, a concentrated alco-
holic beverage results in lower BACs than a dilute one because: 1) it is
absorbed more slowly, and 2) more of the dose is metabolized by FPM,
and thus less alcohol is absorbed.

76. Plasma Membrane Fatty Acid Binding Protein Is Identical to Mi-
tochondrial Aspartate Aminotransferase. DD Stump, S-L Zhou, and
PD Berk. Departments of Medicine and Biochemistry, Mount Sinai
School of Medicine, New York, NY 10029.

Studies over the past decade have demonstrated that hepatocellular
uptake of long chain free fatty acids (FFA) has the kinetic features of
a facilitated transport process. A 43 kDa plasma membrane fatty acid
binding protein (FABPpm), initially isolated at low jdelds by oleate-
agarose affinity chromatography (FABPfl); PNAS 82:4) has been pro-
posed as the putative FFA transporter. Later methods for isolating
FABPpm by salt extraction and HPLC (FABPf21) gave higher yields of
purer product; the isolated protein was found to be related structurally
and immunologically to the mitochondrial isoform of aspartate ami-
notransferase (mAspAT), a soluble mitochondrial matrix protein with
no known role in FFA metabolism (PNAS 87:3484). This conclusion
has been challenged in reports claiming that FABPfl] and FABP[21
were not the same protein (Mol Cell Biochem 98;191).

To resolve this question, samples of FABP(1], FABP[2], and
mAspAT were freshly isolated from rat liver, purified to homogeneity,
and used to prepare polyclonal antisera in rabbits. The three protein
preparations had an identical molecular size (43 kDa) on SDS-PAGE
and gel permeation HPLC, had identical retention characteristics on
four additional HPLC columns, and exhibited a similar pattern of mul-
tiple charge isoforms, pi ca 9.1, by isoelectric focusing. Electrophoresis
under non-denaturing conditions also revealed no differences. Amino
acid compositions of the preparations were similar within the resolu-
tion of the method, and N-terminal amino acid sequences were identi-

cal. FABPdl bound FFA avidly. As initially isolated, FABP(21 and
mAspAT bound FFA poorly; however, after retention on and elution
from oleate agarose columns, both preparations bound FFA with affin-
ity similar to FABPfl 1. All three preparations were subjected to tryptic
digestion, and the resulting peptides separated by reversed phase
HPLC.

No differences in tryptic peptide patterns were observed among the
three proteins. All three proteins produced a line of immunologic iden-
tity in agar gel diffusion studies employing the freshly prepared anti-
sera. On Western blots, each reacted with antisera raised against the
others.

We conclude that FABPfll, FABPf2] and mAspAT are indistin-
guishable. Transfection studies in 3T3 fibroblasts (Isola et al, this
meeting) support this conclusion. How a single mature protein can be
synthesized so that its distribution to two separate cellular locations is
differentially regulated remains to be determined. Preliminary studies
suggest this may occur at the transcriptional level by differential splic-
ing.

77. Alcoholism Is Associated with Hepatitis C (HCV) but not Hepatitis
B at the Bronx VA Medical Center. A. Waraich, A. Rosman, K. Galvia,
R. Williams, F. Paronetto, C. Lieber. Alcohol Research and Treatment
Center, Section of Liver Disease, and Pathology and Laboratory Med-
icine Service, Bronx VA Medical Center and Mount Sinai School of
Medicine, Bronx, New York.

Previous studies have suggested an association of viral hepatitis with
alcoholism. However, the role of confounding risk factors (intravenous
drug use, blood transfusions) has not been adequately excluded.

We therefore performed HBsAg, anti-HBc, anti-HBs, and anti-
HCV (by second generation ELISA) tests in 150 consecutive alcoholics
admitted for detoxification and 111 randomly selected patients attend-
ing a general medical clinic who were also screened for alcoholism
using the brief MAST and CAGE questionnaires. Of the 150 consecu-
tive alcoholics, 50 had a history of intravenous drug abuse (IVDA), 11
had received blood transfusions (BT), and 89 denied any known risk
factor for HCV. Of the 111 patients attending general medical clinic, 21
had a possible history of alcohol abuse, 14 had BT, and 76 denied any
alcohol abuse, IVDA, or BT.

The overall anti-HCV seropositivity in actively drinking alcoholics
was 36% and had a strong association with known risk factors (82% in
IVDA, 27% in BT, contrasting to 11% in alcoholics without known
HCV risk factors, P < 0.001). The overall hepatitis B seropositivity in
alcoholics was 49% and had a strong association with IVDA (86% in
rVDA vs. 46% in BT and 30% in alcoholics without known risk factors,
P < 0.001). Four of the 150 alcoholics (2.7%) were HBsAg positive. The
anti-HCV seropositivity in alcoholics without risk factors was signifi-
cantly greater than in non-alcoholic patients without risk factors (11%
vs. 1%, P = 0.01). In contrast, hepatitis B seropositivity in alcoholics
without risk factors was not significantly different from the non-alco-
holic group (29.5% vs. 30.3%, P > 0.5), Stepwise logistic regression
analysis revealed that alcoholism was a significant risk factors for
hepatitis C (adjusted odds ratio 7.6, 95% confidence interval 1.6-35.5,
P = 0.01) but not for hepatitis B seropositivity (P = 0.8). In alcoholics
without known HCV risk factors, there were no significant differences
between anti-HCV positive vs. anti-HCV negative patients with regard
to age, race, tatoos, and cocaine use. Anti-HCV positive alcoholics had
higher serum admission AST (mean ± SEM, 121 ± 16 vs. 71 ± 6, P <
0.001) and ALT values (91 ± 15 vs. 54 ± 5, P < 0.005) compared to
anti-HCV seronegative alcoholics. However, there were no significant
differences in admission serum GGT values (258 ± 37 vs. 274 ± 59, P
> 0.8) and serum AST/ALT ratio (1.60 ± 0.13 vs. 1.53 ± 0.09, P > 0.6)
between the two groups.

In conclusion, hepatitis C but not hepatitis B seropositivity is in-
creased in actively drinking patients without known risk factors, sug-
gesting that alcoholism, in some way, is a predisposing factor for HCV
infection. In addition, admission serum transaminEises (but not serum
GGT or the AST/ALT ratio) are higher in anti-HCV seropositive alco-
holics than in anti-HCV negative alcoholic patients.

78. Kinetic Characterization of High Affinity, CI" Dependent BSP
Uptake in Isolated Rat Hepatocytes. K Wolfe, DD Stump, LM Isola,
and PD Berk. Departments of Medicine and Biochemistry, Mount Sinai
School of Medicine, New York, NY.

Published studies describe two seemingly distinct sinusoidal upteike
systems for BSP: a "high affinity/low capacity" (reported Vmax 3-4
pmol/min/10® cells; Km < 0.3 jjlM) CI~ dependent system (system A)
and a "low affinity/high capacity" (reported Vmax 3-7 nmol/min/10®

434

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

cells; Km 2-14 ^M) CI" independent system (system B). Both have
been characterized under non-physiologic conditions. System B has
been studied with hepatocyte suspensions in the absence of albumin;
estimated kinetic parameters were subject to errors due to substrate
depletion of =s50'7r during incubation. System A was studied exclu-
sively in monolayer cultures at an unphysiologically low concentration
of BSA; reported kinetic psirameters were therefore subject to pseudo-
facilitation errors.

To better define the kinetic characteristics of system A, we have
studied [''^Sl-BSP uptake by hepatocyte suspensions from a medium
consisting of 600 p-M BSA, either in Hanks' buffer or in a buffer of
similar cation composition in which CI " was replaced by other anions.
BSP:BSA ratios ranged from 0.25:1 to 3:1, with corresponding unbound
BSP concentrations (BSP„) from 0.015 to 3.35 jjlM (JCI 45:281). Incu-
bation volumes were chosen so that uptake depleted the media by
s0.2'7( of incubated substrate. f^^Sl-BSP uptake was determined by
rapid filtration; initial uptake velocities (V,,) were estimated by least
squares fitting of the initial, 20 sec, linear portion of the uptake/time
curve. The relationship between BSP„ and V„ was determined by fit-
ting the data to physiologically relevant functions using the SAAM
program.

Results. At all BSP^ studied, V,, was greater in the presence than
in the absence of Cl " . When the data were fitted to the Michaelis-
Menten equation, as has been conventionally done, Vmax was signif-
icantly reduced and Km significantly increased in the absence, com-
pared to the presence, of Cl " (Vmax = 5.00 ± 0.48 vs 6.86 ± 0.28
nmol/min/10^ cells, p < 0.005; Km = 0.691 ± 0.087 vs 0.405 ± 0.074
M-M, p < 0.01). However, both in presence and absence of Cl " , the best
computer fit consisted of the sum of a saturable plus a linear compo-
nent, which yielded quite different values for the kinetic parameters:
Vmax = 1.77 ± 0.43 vs 2.42 ± 0.14 nmol/min/10*^ cells; Km = 0.207 ±
0.080 vs 0.094 ± 0.012 jjlM, without and with Cr, respectively.

Conclusions. Hepatocytes exhibit a high affinity Cl" dependent
BSP uptake system; its Vmax, however, is similar to that reported for
the low affinity system. Valid comparisons with the low affinity system
require new studies that avoid substrate depletion, pseudofacilitation
and data fitting Eu^ifacts.

79. Differential Expression of the Plasma Membrane Transport Sys-
tem for Long Chain Free Fatty Acids in Obese Zucker Rats. S-L Zhou,
DD Stump, LM Isola, ND Theise, and PD Berk. Departments of Med-
icine, Biochemistry and Pathology, Mount Sinai School of Medicine,
New York, NY 10029.

Uptake of long chain free fatty acids (FFA) has the kinetic features of
facilitated transport in hepatocytes, adipocytes, jejunal enterocytes and
cardiac myocytes, but not in fibroblasts. A plasma membrane fatty acid
binding protein, FABPpm, was proposed as a putative fatty acid trans-
porter. In several experimental settings, changes in cell surface expres-
sion of FABPpm, measured by immunoassay, have correlated with
changes in Vmax for cellular FFA uptake (e.g. JBC 267: 14456). To
examine whether the FFA uptake system was regulated differentially
in different tissues, we examined uptake of f^Hl-oleate by suspensions
of intra-abdominal adipocytes and hepatocytes from homozygous obese
(fa/fa) Zucker rats; heterozygous (Fa/fa) non-obese Zucker or normal
Wistar rats served as controls.

All animals were 6-8 week old males. Studies were conducted in
media containing 500 jjlM BSA and 50-1000 (jlM oleate; the unbound
oleate concentration [0„] ranged from 6.5-430 nM. Uptake was deter-
mined by rapid filtration and initial uptake velocities computed over
the initial 45 sec by least squares. All studies were performed at least
in triplicate. Data were fitted to the Michaelis-Menten equation using
the SAAM program.

Results were:

Hepatocytes

Adipocytes

Obese (fa/fa)
Non-obese (Fa/fa)
Normal Wistar

Vmax

3.0 ± 0.1
2.3 ± 0.1
ND.

75 ± 5
75 ± 3
N.D.

Vmax

60.0 ± 8.5 32.1 ± 7.5
10.3 ± 0.6 14.9 ± 1.9
7.6 ± 0.3 13.1 ± 1.2

only 30% in fa/fa animals compared to Fa/fa controls (p < 0.005); Km
(nM unbound oleate) was unchanged. In adipocytes, the Vmax in fa/fa
animals was increased by 480% vs non-obese Fa/fa (p < 0.005) and
690% vs Wistars (p < 0.001 1; Km's did not differ significantly between

fa/fa and Fa/fa animals (p > 0.05). In H&E sections, hepatocytes from
obese animals were normal in size and appeared to contain only min-
imal fat, although oil red O staining documented substantial lipid ac-
cumulation. By counting number of cells/10 randomly selected hpf in
sections of abdominal fatty tissue, the radius of adipocytes from fa/fa
animals was found to be increased by a mean of 44% compared to Fa/fa
animals; the estimated doubling of surface area, 207%, was insufficient
to explain the increase in oleate uptake Vmax.

Conclusions. These studies demonstrate that regulation of the sat-
urable FFA transport system is tissue specific.

Molecular Medicine

80. Effects of Phospholipid Surface and Mass Transport on Factor X
Activation by Tissue Factor-Factor Vila Complex in a Continuous
Flow Enzyme Reactor. Harry Andree, Paul Contino, Doris Repke, and
Yale Nemerson. Division of Molecular Medicine and Department of
Biochemistry, Mount Sinai School of Medicine, New York, N.Y.

The complex of the treinsmembrane protein tissue factor (TF) and co-
agulation factor Vila is the primary initiator of blood clotting. The
factor X activation rate of this complex w£is studied in the continuous
flow reactor.

Capillary tubes were coated with membranes of 30% phosphati-
dylserine (PS) and 70% phosphatidylcholine (PC) containing TF. The
activation rate of factor X was measured as a function of TF surface
density. Due to the transport rate limited delivery of factor X towards
the surface, saturation was observed.

At a wall shear rate of 1600 s half-maximal conversion of 150
nM factor X was reached at 3 fmoles TF-VIIa/cm^, i.e. =0.04%) of the
surface covered with enzyme. The density that results in half-maximal
saturation increased with substrate concentration and wall shear rate.
The appearance of factor Xa at the outlet of the tube showed a large
delay, explained by high-affinity binding of factor Xa to the phospho-
lipid bilayer on the wall of the tube (Kj = 33 nM). At 70 fmoles TF-
Vlla/cm^, factor Xa was found to cover 80% of the phospholipid surface.
Annexin V, a protein with high-affinity for phospholipids, displaces
proteins from the surface and interferes with membrane mediated
transport. TF-VIIa activity at high TF densities, could not be supressed
by annexin V, which indicates that membrane mediated transport of
factor X towards the enzyme complex is of limited importance. At di-
lute TF surface densities the inhibition increased up to 45%. Experi-
ments at high TF-densities, with TF embedded into pure phosphatidyl-
choline membranes, resulted in a 70% reduction of activity.

This suggests an important role of PS for TF-VIIa activity. In cell
membranes, loss of phospholipid asymmetry results in exposure of PS,
and thus may enhance TF-VIIa activity.

81. Functional Expression of the Skeletal Muscle Ryanodine Recep-
tor. Anne-Marie B. Brillantes, Karol Ondrias, Evgeny Kobrinsky, An-
drew Scott, Barbara E. Ehrlich,* Andrew R. Marks. Molecular Medi-
cine Program, Mount Sinai School of Medicine, NY, NY, *Department
of Medicine, University of Connecticut, Farmington, CT.^

The molecular events leading to activation of the calcium release chan-
nel/ryanodine receptor (RyR) during excitation-contraction coupling in
skeletal muscle remain poorly understood. To gain insight into the
molecular basis of RyR function, we constructed a full-length (—15.2
kb) rabbit skeletal muscle RyR cDNA and functionally expressed it in
Xenopus oocytes.

Functional expression of a calcium release channel was confirmed
by activation of the endogenous Ca^* -sensitive Cl~ channel in re-
sponse to caffeine-induced intracellular calcium release (n = 30). Un-
injected oocytes (n = 25) showed no response to 50 mM CEiffeine. The
565,000 MW cloned expressed RyR (confirmed by ^H-Ryanodine bind-
ing. Northern and Western blot analyses) was purified using sucrose
density gradients and reconstituted in planar lipid bilayers. Single
channel recordings were performed to determine that the properties of
the expressed channel matched those of the native skeletal RyR. It has
been proposed that other molecules may be required for the RyR to
form a functional calcium release channel.

The present study demonstrated that: (1) the RyR alone is suffi-
cient to form a functional skeletal muscle SR calcium release channel;
(2) the ligand binding sites for caffeine, ryanodine, calcium and ruthe-
nium red are encoded by the RyR transcript.

Vol. 60 No. 5

ABSTRACTS

435

82. Calcium Release Channel Expression in Human Heart Failure.
Loewe O. Go, Kushal K. Handa, Tania M. Nanevicz, Adam L. Wilkes,
Billie S. Fyfe, and Andrew R. Marks. Molecular Medicine Program,
Mount Sinai School of Medicine, NY, NY.

The ryanodine receptor (RyR) is the mEyor Ca^ * release channel (CRC)
of cardiac and skeletal muscle activated during excitation-contraction
coupling. The inositol 1,4,5-trisphosphate receptor (IP3R) is the major
CRC of vascular smooth muscle and brain, activated hormonally via
phosphoinositide turnover. In two previous studies we showed that
RyR expression was decreased in end-stage human heart failure (Circ
Res 1992; 71:18), and that the IP3R is expressed in rat cardiac myo-
cytes (JCB 1993; 120:1137).

In the current study we examined CRC expression in failing and
normal (NL) human hearts. RNA isolated from left atrial (LA) and left
ventricular (LV) specimens obtained prospectively from patients un-
dergoing cardiac transplantation (n = 20) and NL controls (n = 4) was
analyzed using northern and slot blots. Quantification was with a phos-
phorimager, and CRC mRNA levels were normalized to 28S ribosomal
RNA levels.

Northern blot analyses of human heart RNA detected a single ~15
kb mRNA with a 580 bp rabbit cardiac RyR cDNA probe, and a distinct
-10 kb mRNA with a 1.7 kb rat aortic smooth muscle IP3R cDNA
probe. Compared to NL, RyR expression in failing hearts was again
decreased in the LV (by 31%), but was increased in the LA (by 40%). In
contrast, IP3R expression was increased in both LA and LV (by 30-
50%).

We conclude that in contrast to the RyR, the IP3R mRNA is up-
regulated in end-stage cardiomyopathy. This up-regulation may be
compensatory by providing an alternative pathway for mobilizing in-
tracellular Ca^ * release.

83. Effects of Adriamycin and Thyroxine on Calcium Release Channel
Expression in a Rat Model of Heart Failure. Kushal K. Handa, Loewe
0. Go, Maria T. Moschella, and Andrew R. Marks. Molecular Medicine
Program, Mount Sinai School of Medicine, New York, NY.

Adriamycin-induced cardiomyopathy is a major toxic side effect which
limits the use of this potent anti-tumor drug. The molecular basis for
this drug-induced cardiomyopathy remains obscure; however, a gener-
alized decrease in cardiac-specific gene expression occurs with adria-
mycin treatment. This is in contrast to thyroxine treatment which
results in generalized increase in gene expression. Cardiac muscle con-
traction is an intricate process in which calcium release channels
(CRCs) play a pivotal role. The entry of calcium through the voltage
dependent calcium channel on the plasma membrane leads to the ac-
tivation of the ryanodine receptor (RyR), the major CRC of cardiac
muscle involved in excitation-contraction coupling. The inositol 1,4,5-
trisphosphate receptor (IP3R) is another intracellular CRC recently
found to be expressed in cardiocytes but whose functional significance
is unclear.

An animal model of heart failure was established using adult rats
by injecting a cardiotoxic dose of adriamycin (8 mg/kg intraperitoneal-
ly). The animals were then sacrificed 3, 6, 9, and 18 days after injection
of adriamycin along with age and sex matched controls. Northern and
slot blot analyses were performed on total RNA extracted from the rat
hearts using the rat IP3R and rabbit cardiac RyR cDNA probes.

Compared to controls (n = 12), RyR expression was markedly de-
creased —50% in the treated hearts (n = 20) at all time points, but
there was a nonsignificant increase in IP3R expression. Preliminary
data showed that thyroxine treatment (4 mg/kg) is associated with a
four-fold increase in RyR expression. Simultaneous treatment with
adriamycin and thyroxine appears to abolish these changes in RyR
expression, but have an additive effect in increasing IP3R expression
two-fold.

These results imply a role of the CRCs in heart failure and suggest
that adriamycin-induced cardiomyopathy is mediated via altered CRC
gene expression.

84. Modulation of Expression and Function of Putative Natural Killer
(NK) Cell Inhibitory Receptors by Host Major Histocompatibility
Complex (MHO Class I Molecules. Franz M. Karlhofer, Rosemarie
Hunziker, David H. Margulies, and Wayne M. Yokoyama. Molecular
Medicine Division, Department of Medicine, Mount Sinai Medical Cen-
ter, New York, NY, & Laboratory of Immunology, NIAID, NIH, Be-
thesda, MD.

The Ly-49 molecule is expressed on a subset (15-20%) of splenic
NKl.l^ cells from C57BL/6 or C57BL/10 (H-2'') mice. The cytolytic

activity of Ly-49 ' NK cells is specifically inhibited by target cell ex-
pression of H-2D'' or an H-2'* class I molecule, consistent with the
possibility that Ly-49 directly engages these MHC class I molecules.

To determine the influence of host MHC molecules on Ly-49 ex-
pression, we examined splenic NKl.l ^ cells from MHC-congenic and
transgenic strains of the BIO or B6 background by two-color flow cy-
tometry.

Ly-49* NK cells were undetectable in F, hybrid and C57BL/10
MHC-congenic mice expressing the MHC class I molecules, D** or D'',
and in a C57BL/6 strain transgenic for membrane bound D"*. In con-
trast, Ly-49* NK cells were detectable in two strains transgenic for
soluble forms of D''. Ly-49* NK cells are also absent in B10.D2"*"''
(al''a2'^'a3') mice. This demonstrates that the absence of Ly-49* NK
cells depends on specific expression of al/a2 domains of membrane
bound D''. Ly-49* NK cells are present in BIO.AKM (K'^,D''), and
BIO S (K'',D'') mice and in C57BL/6 mice transgenic for a mutation in
the Pj-microglobulin gene, confirming that Ly-49 expression is not
restricted to the H-2'' haplotype and suggesting that expression is not
dependent on normal H-2'' expression.

Although these studies suggest that Ly-49 * NK cells may be per-
manently deleted by negative selection, analogous to T cell clonal de-
letion, Ly-49* NK cells can be cultivated in vitro from D''-bearing
mice, indicating that host MHC class I molecules modulate the expres-
sion of Ly-49 in a manner different fi-om clonal selection. However,
these Ly-49 * NK cells are capable of killing a D"'-expressing transfec-
tant, suggesting that the function of Ly-49 on these cells is also altered.
Taken together, these data confirm our in vitro studies of Ly-49 spec-
ificity for MHC class I molecules. Moreover, these studies suggest that
MHC class I molecules play important roles in the development of NK
cells and establishment of the NK cell "repertoire."

85. Functional and Mutational Analysis of the Drosophila Kruppel
Transcriptional Repressor. J. D. Licht, M. A. English, J. C. Reddy, M.
Ro, R. Shaknovich, M. Grossel, W. Hanna and U. Hansen. Division of
Molecular Medicine, Mount Sinai School of Medicine, New York, NY,
and Dana-Farber Cancer Institute, Harvard Medical School, Bos-
ton, MA.

The Drosophila Kriippel (Kr) protein is a DNA-binding repressor of
transcription in mammalian cells. (Nature 346: 76-9, 1990). The tran-
scriptional repression and DNA-binding activities of the Kr protein can
be dissociated and the repression function otKr can be transferred to a
heterologous DNA-binding protein, the lac repressor. Fusion genes be-
tween portions of the N-terminal region of the Kr protein and the lac
repressor were assayed in vivo for their ability to repress transcription
from a reporter gene containing lac operators upstream of the herpes
thymidine kinase (tk) promoter.

We thus localized a repression region between amino acids 60 and
90 of the Kr protein. This portion of the Kr protein is predicted to form
an alpha-helix with several hydrophobic faces.

However mutant proteins deleted for either the N-terminal or
C-terminal portion of the protein were still able to repress transcrip-
tion from a reporter gene containing Kr binding sites, suggesting the
presence of redundant repression domains. Supporting this hypothesis,
a fusion between the lac repressor and amino acids 402-502 of Kr
repressed transcription from a lac operator containing reporter gene. In
accordance with our prediction that alanine-rich regions of other
Drosophila transcriptional factors would mediate transcriptional re-
pression, a lac repressor even-skipped fusion was six-times more potent
a transcriptional repressor than lac-Kr. The presence of a stretch of 11
alanines in a row as well as a proline-rich region in the lac-eve protein
was required for repression.

Kr represses transcription when bound at kilobase distances up-
stream or downstream from the start site of tk transcription, suggest-
ing function through protein-protein interactions with factors bound to
the tk promoter. Low amounts of Kr co-transfected with a Kr binding
site containing reporter gene do not lead to activation and a monotonic
decrease in transcription is observed with increasing amounts of Kr
expression plasmid. We examined the ability of the Kr protein to re-
press transcription activated by different GALA fusion activators. Kr
selectively repressed transcription activated by a GAL4-Spl, a glu-
tamine-rich activator protein but could not repress transcription me-
diated by the acidic activator GALA. Correlating with this, we found
that Kr and Spl proteins can associate in vitro. We currently believe
that Kr functions to quench activation by glutamine-rich activation
domains, perhaps by a direct protein-protein interaction, rather than
by interfering with some component of the basal transcriptional ma-
chinery required for all transcriptional activation events.

436

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

86. PLZF, a Myeloid Specific Transcription Factor Rearranged with
The Retinoic Acid Receptor in Acute Leukemia. J. D. Licht, A. Chen,
Y, Wu, J. Li, M. A. English, R. Shaknovich, W. Miller, Jr., A. Scott, C.
Willman, and S. Waxman. Divisions of Molecular Medicine and Med-
ical Oncology, Mount Sinai School of Medicine, New York, NY, Memo-
rial Sloan Kettering Cancer Center, New York, NY, and University of
New Mexico, Albuquerque, NM.

Molecular analysis of a patient with acute promyelocytic leukemia
revealed a translocation t(ll;17) and a rearrangement between the
gene encoding the retinoic acid receptor a (RARa) and novel gene
called PLZF (Promyelocytic Leukemia Zinc Finger) (Z. Chen et al,
1993, S. Chen et al., 1993). The rearrangement yields two fusion pro-
teins, one fusing the A domain, a transcriptional activation domain to
7 zinc fingers of PLZF. The other fusion protein contains the amino
terminal region of PLZF including two zinc fingers and the B-F do-
mains of RARa, including the DNA binding domain and ligand binding
domain of the receptor. We have subsequently identified two additional
patients with PLZF-RARa rearrangement and one of these patients
had a novel site of rearrangement in the PLZF gene. PLZF is expressed
in myeloid HL60 and NB4 cell lines, but not erythroid or lymphoid cell
lines or the monocytoid U937 cell line. PLZF is expressed in the bone
marrow of normal subjects as well as those with acute leukemia. In
peripheral blood, PLZF is expressed in tissue macrophages but not in
mature granulocytes. PLZF expression is down regulated during gran-
ulocytic differentiation of HL60 cells induced by retinoic acid £ind
monocytoid differentiation induced by vitamin D3. Intriguingly, PLZF
appears to be expressed in prostate cancer specimens but not in spec-
imens from patients with benign prostatic hypertrophy. PLZF encodes
a 673 amino acid protein with a predicted molecular weight of 74kd,
containing 9 putative DNA-binding zinc finger motifs. When expressed
in non-hematopoetic cells, the PLZF cDNA produces a ~80kd nucleeu"
localized protein. The DNA binding activity and possible target genes
of PLZF are under investigation.

To determine the transcriptional effector function of PLZF we
fused the amino terminal portion of the protein to the DNA binding
domain of the yeast GALA protein.

This chimeric protein was a potent repressor of transcription of a
test promoter containing GALA DNA binding sites upstream of the
herpes simplex virus thymidine kinase promoter.

This result suggests a possible mechanism for leukemogenesis by
RAR-PLZF fusion; PLZF is normally a repressor of its target genes, but
when fused to the A domain of the RAR becomes a constitutive acti-
vator. This may also explain why, in contrast to patients with t(15;17)
APL the index patient as well as four other patients with the PLZF-
RAR rearrangement failed to respond to retinoic acid or conventional
chemotherapy.

87. Structural Organization of the Human Gene Encoding Nuclear
Lamin A and Nuclear Lamin C. F. Lin and H. J. Worman. Division of
Molecular Medicine, Dept. of Medicine and Brookdale Center for Mo-
lecular Biology, Mount Sinai School of Medicine, New York, NY.

The nuclear lamins are intermediate filament proteins of the nuclear
envelope lamina. In most terminally differentiated mammalian cells,
three major lamins termed lamin A, lamin B and lamin C are ex-
pressed. Lamin B is present in all cells whereas lamins A and C are
absent from some undifferentiated and cancer cells. Lamins A and C
are identical for the first 566 amino acids and differ in their carboxyl-
termini. Lamin A is synthesized as a precursor called pre-lamin A
which is isoprenylated. The last 18 carboxyl-terminal amino acids of
pre-lamin A are proteoljrtically removed to generate lamin A. To better
understand the above findings, as well as the evolution of the inter-
mediate filament multigene family, we have determined the structural
organization of the human gene that encodes nuclear lamins A and C.

Sequencing and restriction mapping show that the coding region
spans approximately 24 kilobases (kb). The 5' proximal promoter re-
gion contains several GC-rich stretches, a CCAAT-box and a TATA-
like element of sequence TATTA. The lamin PJC gene contains 12
exons. Exon 1 codes for the amino-terminal head domain and the first
part of the central rod domain common to lamins A and C. Exons 2-6
code for the remainder of the rod domain. Exons 7-9 code for the
carboxyl-terminal tail domain sequences common to both lamin A and
C; the sequence encoding the nuclear localization signal is contained in
exon 7. Alternative splicing within exon 10 gives rise to two different
mRNAs that code for pre-lamin A and lamin C. Exons 11 and 12 are
lamin A-specific exons with the sequence coding for the CAAX-box
isoprenylation motif in exon 12. As a consequence of alternative RNA

splicing within exon 10, two proteins are generated, only one of which,
pre-lamin A, can be modified by isoprenylation. The intron positions in
the human lamin PJC gene are generally conserved in the previously
characterized genes for Xenopus lamin LIII and mouse lamin B2, but
differ ft-om those in a Drosophila lamin gene. In the regions coding for
the central rod domains, the intron positions are also conserved when
compared to the intron positions in the genes for most cytoplasmic
intermediate filament proteins except those for nestin and neurofila-
ments.

Analysis of the intron positions in these genes supports the hy-
pothesis that the nuclear lamins and other intermediate filament pro-
teins arose from a common ancestor. The lamin A/C gene is also the
first example of how two different protein isoforms that differ in their
capacity to be isoprenylated are generated by alternative RNA splic-
ing.

88. The Characterization of Tissue Factor in Human Coronary Ather-
oma. Jonathan D. Marmur, Billie Fyfe, Samin K. Sharma, John A.
Ambrose, Marc Habert, Sarah Eisenberg, Alsin Lotvin, Douglas Israel,
Sabino R. Torre, Yale Nemerson, Mark B. Taubman. Mt. Sinai Medical
Center, NY, NY.

Tissue Factor (TF) initiates the predominant pathway of blood coagu-
lation in vivo and has been detected in atherosclerotic plaque derived
ft-om carotid artery and saphenous vein grafts.

To characterize the expression of TF in coronary atheroma, immu-
nohistochemistry with a monoclonal anti-human TF antibody was per-
formed on frozen sections of 24 lesions from 24 patients treated with
directional coronary atherectomy.

Positive staining for TF antigen was demonstrated in specimens
derived from 8 (33%) of the lesions. TF antigen weis distributed in a
non-diffuse pattern, with small foci detectable predominsintly in areas
of acellular sclerotic plaque. Areas of intimal hyperplasia were not
associated with positive TF staining. Differences between patients
with positive TF stain (TF*) versus patients with negative TF stain
(TF ~ ) were analyzed using Fisher's exact test. Lesion morphology was
defined as complex when an ulceration, irregular borders, or a filling
defect was present on baseline angiography. All patients had 6 month
clinical follow-up. Restenosis was defined as recurrence of symptoms
with positive exercise perfusion imaging, or >50% diameter stenosis on
angiography.

TF-H (n = 8)

TF- (n = 16)

unstable emgina

(rest or post-MI pain) 7 (88%) 14 (88%)

thrombus on histology

H&E staining 6 (75%) p = 0.08 5 (31%)

complex angiographic

morphology 6 (75%) p = 0.03 4 (25%)

restenosis

3 (38%) p = 0.36 3 (19%)

The association of TF antigen with a complex eingiographic lesion
morphology and the trend toward an association between TF antigen
and histologic thrombus support a role for TF in the thrombotic events
associated with coronary atherosclerosis.

89. Detection and Quantification of Tissue Factor Activity in Coro-
nary Atherectomy Specimens. Jonathein D. Marmur, ThiruvikrEimsm
V. Singanallore, Billie Fyfe, Samin Sharma, Arabinda Guha, Yale
Nemerson. Depts. of Medicine and Pathology, Mt. Sinai Med. Ctr., NY.

Tissue factor (TF) is a membrane-boimd clotting factor that initiates
the major pathway of coagulation in vivo by binding factor VII (FVII).
The TF;FVII complex converts factor X (FX) to its active form, Xa,
thereby activating the clotting cascade. We have detected TF antigen
in =V3 of coronary atherectomy specimens using a monoclonal anti-TF
antibody.

To detect and quantify TF activity, humsm coronary atherectomy
specimens (n = 15) were incubated with an assay mix (AM) containing
inactive FX (100 nM), FVII (1 nM), and calcium (5 mM) at 37°C in
microliter plate wells. In these conditions, the amount of Xa generation
is determined by the quantity of bioavailable TF. To quantify Xa gen-
eration, a Xa-sensitive chromogen (ILeu-Pro-Arg-pNA) was added to
the wells that produces a color change measured at 405 nM in an

Vol. 60 No. 5

ABSTRACTS

437

ELISA counter. TF activity in wells containing AM alone, and AM and
plaque (AM + P) were read after various periods of incubation.

TF activity (mean ± SD) is expressed in optical density units (milli
OD) on the vertical axis:

40 T

20--

..I

0.5 2.5
AM

19 .0.5 2.5 19

HOURS AM-t-P

Factor Xa generation was time-dependent and detectable at high levels
in wells containing atherosclerotic plaque but not in wells containing
AM alone. In AM alone wells (n = 15), the optical density was always
<3 milli OD. At 2.5 hrs, 13 of the 15 (87%) atherectomy specimens
generated >3.0 milli OD. Procoagulant activity at 0.5 hrs was elimi-
nated by pre-incubation of the plaque with anti-TF antibody.

Conclusions. TF activity is detectable and quantifiable in coronary
atheroma. The activity assay appears to be more sensitive than immu-
nohistochemistry for the detection of TF (87% versus 33%). TF may
constitute an important procoagulant molecule within the coronary
atherosclerotic plaque.

90. Enhanced Expression of Tissue Factor in the Vessel Wall in Pre-
viously Injured versus Normal Artery Following Balloon Dilatation.

Jonathan D. Marmur, Maria Rossikhina, Nicholeis J. Gargiulo, Ara-
binda Guha, Milton Mendlowitz, Yale Nemerson, Mark B. Taubman.
Mt. Sinai Hosp., NY.

The initiation of thrombosis after vascular injury occurs within min-
utes. We have recently demonstrated that tissue factor (TF) mRNA and
activity are induced in normal rat aortic media at =2 hrs after balloon
injury, suggesting that acute activation of the TF gene may play a role
in the thrombus propagation. In the rabbit aorta balloon injury model
it has been shown that in comparison to balloon dilatation of a normal
artery, balloon dilatation of a previously injured artery that has devel-
oped a neointima is associated with a marked increase in fibrin depo-
sition.

To examine the role of TF in the re-injury model, rat aortas were
subjected to a second balloon injury 2 weeks after the first (n = 3
rats/time point). Rat aortic TF mRNA was examined by Northern blot
analysis and in-situ hybridization. TF activity was measured using a
recently developed colorometric assay of factor Xa generation and is
expressed in arbitrary TF units/mg of total protein (U/mg; mean ± SD).
Total protein was determined using the Bradford reagent.

After a single injury, rat aortic TF mRNA and activity were ele-
vated at 2 hours compared with control (101 ± 21 vs 29 ± 10 U/mg),
and returned to baseline by 24 hours. At 2 weeks, no increase in TF
mRNA or activity was seen and intimal hyperplasia was consistently
present. After a second injury TF activity was induced within 30 min-
utes (88 ± 14 U/mg) and peaked at 1 hr to higher levels than those seen
after a single injury (202 ± 47 U/mg). Mural thrombus was detectable
on light microscopy after double injury but not after single injury.

In previously injured vessels with a neointima, induction of TF
expression in arterial media occurs more rapidly and to a greater de-
gree than in normal artery. Injury to diseased vessels, such as that seen
during coronary angioplasty, may involve an enhanced expression of
TF that participates in thrombus initiation.

91. Intracoronary Thrombin Generation and Activity During Angio-
plasty (PTCA). Jonathan D. Marmur, Samin K. Sharma, Neda
Khaghan, Fred Resnick, Kenneth Bauer, Sabino R. Torre, Robert D.
Rosenberg, John A. Ambrose. Mt. Sinai School of Medicine, New York,
NY; Beth Israel Hospital, Heirvard University, Boston, MA.

Initiation of the clotting cascade ultimately results in the conversion of
fibrinogen to fibrin and of prothrombin to thrombin. Fibrinopeptide A
(FPA) and prothrombin fragment F1 + 2 •F1 + 2) are released by these
processes, respectively, and have been used as biochemical markers of
coagulation cascade activity. We have previously shown that in sys-
temically heparinized patients, levels of FPA and Fn.2 "lay be accu-
rately determined in samples withdrawn through a PTCA catheter.
To monitor the activation of thrombosis during PTCA, we collected

blood samples before (Pre-PTCA) and 10 minutes after the final balloon
inflation (Post-PTCA) in 39 patients. Samples were taken simulta-
neously from a femoral venous sheath (VN) and from the PTCA cath-
eter (AR), which was placed proximal and distal to the lesion for Pre-
and Post-PTCA sampling, respectively. Plasma levels (mean ± SD)
were determined by radioimmunoassay and compared using a paired,
two-tailed t-test.

Pre-PTCA
Post-PTCA

FPA (ng/mL)

5.22 ± 3.49} p < 0.001
4.52 ± 4.86}

3.01 ± 2.59} p = 0.19
4.13 ± 4.06}

Fl -I- 2 (nmol/L)

0.804 ± 0.677} p = 0.56
0.853 ± 0.710}

0.771 ± 0.395} p = 0.26
0.936 ± 0.866}

In the setting of decreasing systemic levels of FPA, as demon-
strated by the results from venous sampling, coronary blood FPA levels
show an increasing trend. PTCA increases thrombin activity within
the coronary artery at the site of balloon injury, despite the systemic
effects of heparin.

92. FK-506 Binding Protein Is Tightly Bound to the Ryanodine Re-
ceptor in the Heart. Roxana Mehran, Anne-Marie Brillantes, T. Ja-
yaraman, Karol Ondrias and Andrew R. Marks. Department of Medi-
cine, Molecular Medicine Program, Mount Sinai School of Medicine,
New York, N.Y.

The calcium (Ca) release from sarcoplasmic reticulum (SR) is com-
prised of four ryanodine receptors (RyR). The RyR is activated during
excitation-contraction coupling in cardiac and skeletal muscle. We
have previously shown that FK-506 binding protein (FKBP12) is phys-
ically associated with RYR isolated from rabbit skeletal muscle SR in
a molar ratio of one FKBP12 to two RyRs.

Northern blot analysis using an FKBP12 cDNA probe showed that
the 1.6 kb mRNA is expressed in skeletal and CEirdiac muscle. We
cloned the full length human cardiac FKBP cDNA from a human ven-
tricular \gGT10 cDNA library.

Sequence analysis demonstrated that the FKBP12 expressed in
human heart is encoded by the same gene as the human T lymphocyte
FKBP12. Immunoprecipitation of cardiac RYR using an anti-FKBP12
antibody demonstrated a physical association between the two mole-
cules. Western blot analysis using an antibody directed against the
amino terminus of FKBP12 showed that FKBP12 was present in skel-
etal and cardiac SR and in highly purified RyR preparations. Addition
of recombinant FKBP12 to cloned expressed skeletal muscle RYR re-
constituted in planar lipid bilayers demonstrated that FKBP12 stabi-
lizes the cloned expressed RyR and increases channel opening.

Thus, we have shown that the Ca release channel of cardiac and
skeletal muscle consists of a complex which includes four of the 565,000
Da RyRs and two FKBP12 (12,000 Da each). FKBP12 appears to opti-
mize the function of the SR Ca release channel involved in EC coupling.

93. Antisense Oligonucleotide Inhibition of Tissue Factor-Induced Co-
agulation. Steven A. Nichtberger, Claire-Lise Rosenfield, Yale Nem-
erson, Mark B. Taubman. Divisions of Cardiology, Molecular Medicine
and Thrombosis. Mt. Sinai School of Medicine, NY, NY.'''

Tissue factor (TF) is a transmembrane glycoprotein that is the primary
initiator of coagulation. We have reported that TF mRNA and activity
are highly regulated in cultured vascular smooth muscle cells (VSMC)
by growth factors, a clotting factor, and in rat aortic media after bal-
loon injury. These observations suggest that induction of TF may be
importEint in thrombosis associated with myocardial infarction and cor-
onary angioplasty.

As a first step toward elucidating the role of TF induction in vas-
cular injury, we have examined the effect of antisense oligonucleotides
(AS) on TF expression in cultured rat aortic VSMC.

TF mRNA (assayed by Northern blot analysis) and activity (as-
sayed in 96 well dishes by a chromogenic assay that measures the
conversion of factor X to Xa) were barely detectable in quiescent VSMC
(grown in serum-free medium (DME/BSA) for 48 h). Following stimu-
lation with 10% calf serum (CS), levels of TF mRNA increased at 15
min, peaked at 60-90 min (=3000 copies/cell), and returned to baseline
by 3 h. TF activity increased at 30 min, peaked at 2-8 h (=5- 10-fold),
and returned to baseline by 24 h. AS flanking the initiation ATG (ASl)
blocked the serum-induced rise in TF activity (>85% inhibition).

438

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

200.00 -r

Tlssue Factor Activity

4 hours after serum treatment

DME/BSA 1 0% CS

CS/S

CS/AS1 CS/AS2

4 h incubation with ASl prior to serum treatment was required for
inhibition. Maximal inhibition was seen at 20 jiM ASl and the effect
persisted for 24 h. The corresponding sense oligonucleotide(S) or an AS
containing a single base mismatch (AS2) failed to inhibit TF expres-
sion.

We conclude that ASl treatment is effective in blocking coagula-
tion mediated by TF, in spite of its expression at very high levels, in
cultured VSMC.

94. Autoantibodies Against Nuclear Pore Membrane Glycoprotein
GP210 in Patients with Primary Biliary Cirrhosis Recognize an
Epitope with Sequence Homology to Bacterial Polypeptides. R. E.

Nickowitz and H. J. Worman. Division of Molecular Medicine, Dept. of
Medicine and Brookdale Center for Molecular Biology, Mount Sinai
School of Medicine, New York, NY.*

Patients with primary biliary cirrhosis (PBC) have autoantibodies
against gp210, the major glycoprotein of the nuclear pore membrane.
Gp210 is composed of a 1808 amino acid amino-terminal domain that is
located in the perinuclear space, a single transmembrane segment and
a 58 amino acid carboxyl-terminal domain that is located on the nu-
cleo-cytoplasmic side of the nuclear pore membrane. We have deter-
mined the epitope(s) of gp210 that is recognized by autoantibodies from
patients with PBC.

Twelve glutathione-S-transferase fusion proteins containing dif-
ferent regions of gp210 were expressed using the prokaryotic expres-
sion vector pGEX-2T and purified by glutathione-Sepharose chroma-
tography. The fusion proteins were used in immunoblotting
experiments with serum from 25 patients with PBC that had autoanti-
bodies against gp210.

None of the autoantibodies recognized the amino-terminal domain
of gp210 located in the perinuclear space. All 25 autoantibodies recog-
nized the 58 amino acid nucleocytoplasmic domain. Immunoblotting of
fusion proteins that contained different regions of the nucleocytoplas-
mic domain showed that autoantibodies from all 25 patients only rec-
ognized fusion proteins that contained the sequence NH2-PPSGLWS-
PAY. When compared to available protein sequences, homology was
found between the recognized epitope of gp210 and a sequence common
to the protein products of the E. coli mutY gene and the S. typhimurium
mutB gene:

gp2 10 epitope: PPSGLWSPAY
!!!!!! !
E. coll mutY gene product: LLAQRPPSGLWGGLYCFP

These findings demonstrate that autoantibodies against gp210 in
patients with PBC only recognize a restricted region in the protein's
nucleo-cytoplasmic domain. The epitope is homologous to a portion of
the E. coli mutY gene product suggesting that these autoantibodies
may arise by molecular mimicry of bacterial antigens. As homologies
between the E. coli and human E2 subunits of pyruvate dehydrogenase
have also been demonstrated, the present data again raise the provoc-
ative question: is there evidence for molecular mimicry in PBC?

95. Autoantibodies Against Recombinant Integral Membrane Proteins
of the Nuclear Envelope in Patients with Primary Biliary Cirrhosis.

R. E. Nickowitz, R. W. Wozniak, F. Schaffner and H. J. Worman. Di-
vision of Molecular Medicine and Division of Liver Diseases, Dept. of
Medicine and Brookdale Center for Molecular Biology, Mount Sinai
School of Medicine, New York, NY and Laboratory of Cell Biology, The
Rockefeller University, New York, NY.

An assay utilizing recombinant polypeptides was designed to detect
serum autoantibodies against two integral membrane proteins of the
nuclear envelope, gp210 and the lamin B receptor. The data were an-
alyzed to determine the significance of these autoantibodies in the
diagnosis of primary biliary cirrhosis (PBC).

Serum samples were obtained from 159 patients with PBC and 46
controls with viral hepatitis, autoimmune hepatitis, or other diseases

from the inpatient and outpatient services of The Mount Sinai Hospi-
tal. Recombinant fusion proteins of nuclear pore membrane glycopro-
tein gp210 and inner nuclear membrane protein lamin B receptor were
expressed using the pGEX-2T vector.

Autoantibodies against recombinant gp210 were detected in 15 of
159 patients with PBC and 0 of 46 individuals who were either healthy
or had other diseases. Antibodies against recombinant lamin B recep-
tor were detected in 2 patients with PBC and 0 control individuals. The
presence of autoantibodies against these proteins had a sensitivity of
11% and specificity of 100% for the diagnosis of PBC. Autoantibodies
against gp210 were present in 4 of 19 (21%) patients with PBC who did
not have detectable antimitochondrial antibodies. Patients with PBC
and serum antibodies against gp210 had a significantly higher inci-
dence of associated rheumatoid arthritis than patients without these
antibodies. The two groups did not otherwise differ in age, sex, routine
laboratory values, serum IgM concentrations or incidences of associ-
ated diseases other than arthritis.

The results show that autoantibodies against gp210 and the lamin
B receptor, integral membrane proteins of the nuclear envelope, are
present in 11% of patients with PBC and can be unequivocally identi-
fied by an assay utilizing recombinant polypeptides. These autoanti-
bodies have a specificity of 100% for the diagnosis of PBC and may be
particularly useful in diagnosing patients without detectable antimi-
tochondrial antibodies. They may also identify a subgroup of patients
that has an increased incidence of associated rheumatoid arthritis.

96. Identification and Characterization of Novel Nuclear Envelope
Proteins Using Autoantibodies from Patients with Autoimmune Liver
Diseases. C. M. Noyer, R. E. Nickowitz and H. J. Worman. Division of
Molecular Medicine, Dept. of Medicine and Brookdale Center for Mo-
lecular Biology, Mount Sinai School of Medicine, New York, NY.

Autoantibodies from patients with primary biliary cirrhosis (PBC) and
autoimmune hepatitis (AH) frequently recognize proteins of the nu-
clear envelope. Autoantibodies from patients with AH generally rec-
ognize the nuclear lamins and those from patients with PBC recognize
gp210 and the lamin B receptor, integral proteins of the inner nuclear
membrane.

We have screened serum samples by immunofluorescence micros-
copy for autoantibodies that recognize the nuclear envelope.

Autoantibodies from 47 of 159 patients with PBC and 6 of 14 pa-
tients with AH gave nuclear rim fluorescence when examined by im-
munofluorescence microscopy, suggesting nuclear envelope reactivity.
Of the 47 samples from patients with PBC, 17 recognized integral
membrane proteins gp210 and lamin B receptor and 4 recognized
lamins. Of the 6 samples from patients with AH, 3 recognized nuclear
lamins. Autoantibodies from one patient with AH reacted with a pro-
tein of approximately 60 kDa on immunoblots of proteins fi-om rat liver
nuclear envelopes that was different from nuclear lamins, lamin B
receptor or other previously characterized antigens. In the samples
from the patients with PBC, two contained autoantibodies that recog-
nized a protein in rat liver nuclear envelopes with an apparent molec-
ular mass of 47 kDa. One sample contained autoantibodies against a
protein of apparent molecular mass 63 kDa on immunoblots of rat liver
nuclear envelope proteins. Autoantibodies purified from these serum
samples by affinity chromatography against the recognized proteins
labeled the nuclear periphery when examined by immunofluorescence
microscopy. Both of these apparently novel protein antigens were not
extracted from nuclear envelope preparations with 1 M NaCl but were
extractable with 8 M urea.

These results confirm that autoantibodies from numerous patients
with autoimmune liver diseases recognize proteins of the nuclear en-
velope. Several of these patients contain autoantibodies that recognize
novel proteins of this organelle that are presently being characterized
in greater detail.

97. Transcriptional Activation and Repression by the Wilms' Tumor
(WTl) Gene Product. J. C. Reddy, J. C. Morris, M. A. English, D. A.
Haber, X. N. Luo, G. Atweh and J. D. Licht. Divisions of Molecular
Medicine, Hematology and Neoplastic Diseases, Mount Sinai Medical
Center, New York, NY 10029 and Massachusetts General Hospital
Cancer Center, Charlestown, MA 02114.

The WT-1 gene product binds to the DNA sequence 5'-GCGGGGGCG-
3', also recognized by the product of the EGR-1 and EGR-2 genes and
was previously shown to be a transcriptional repressor. (Madden et. al.,
1991; Drummond et. al., 1992) In contrast with these results, we found
that the WTl gene product, when expressed in African green monkey
kidney (CV-1) cells and early passage NIH 3T3 cells, activated tran-

Vol. 60 No. 5

ABSTRACTS

439

scription from a reporter gene containing EGR-1 binding sites up to
90-fold.

Transcriptional activation was dose dependent in both cell lines
and did not plateau or decrease at high input levels of plasmid, sug-
gesting that the transcriptional machinery for activation by WT-1 is
not limiting in the cell. An alternatively spliced form of the WTl gene
WTKB) which inserts a serine rich amino acid segment was a more
potent activator than WTl(A), lacking the insert. WTl forms C and D
which contain a three amino acid insert ( -I- KTS) between the third and
fourth zinc fingers of the protein are known not to bind to the EGR site
and did not affect transcription from the reporter gene itself but sig-
nificantly blocked the ability of WTKAI to activate transcription.
WT(AR), (Haber et. al., 1991) a naturally occurring mutant, also pre-
vented transcriptional activation by WT-1 suggesting a dominant neg-
ative activity on the action of the wild-type protein.

These data suggest the formation of an inactive heterodimer be-
tween alternatively spliced forms of WTl or mutant and wild type
forms as a mode to regulate the net transcriptional activity of WTl.

We also studied the ability of WTl to affect transcription of the
promoter of the P18 gene, which as is the case for WTl is highly ex-
pressed in leukemic blast cells. WTl forms A and B activated this
promoter while forms C and D repressed the promoter. A deletion mu-
tant of the promoter, which eliminated GC rich sequences downstream
of the start site of transcription was still activated by forms A and B but
could no longer be repressed by C and D.

This suggests the possibility of ai + KTS) form specific element in
the P18 promoter and also supports the notion that cis-acting elements
both 5' and 3' to the start site of transcription are required for WTl
protein to repress transcription.

98. Structural Organization and Expression of the Human Gene En-
coding the Inner Nuclear Membrane Lamin B Receptor. E. Schuler, F.
Lin and H. J. Worman. Division of Molecular Medicine, Dept. of Med-
icine £Lnd Brookdale Center for Molecular Biology, Mount Sinai School
of Medicine, New York, NY.

The lamin B receptor (LBR) is an integral membrane protein of the
inner nuclear membrane that was originally characterized in birds and
subsequent in humans. It has been proposed that LBR anchors the
nuclear lamina to the inner nuclear membrane by binding to lamin B
in interphase and targets vesicular remnants of the inner nuclear
membrane to decondensing chromosomes at the end of mitosis to ini-
tiate nuclear envelope reassembly. LBR has a basic nucleoplasmic
emiino-terminal domain followed by eight putative transmembrane
segments and displays no overall homology to other proteins of known
sequence. We have now determined the structural organization of the
human gene that codes for this unique nuclear envelope protein.

Sequencing and restriction mapping of overlapping genomic clones
show that the human LBR gene spans more than 50 kilobases (kb). One
transcription start site is located 3 kb 5' to the initiation AUG. The
RNA segment in this region that is spliced from most transcripts may
contain an alternative promoter and transcription start site. The cod-
ing region of the LBR gene contains 1 1 exons. Three exons encode the
nucleoplasmic amino-terminal domain and eight exons encode that pu-
tative transmembrane segments. An intron separates the portion of the
gene encoding the amino-terminal domain from the portion encoding
the transmembrane segments. A yeast gene has been identified coding
for a protein that is homologous to the transmembrane portion of LBR
but lacking the amino-terminal domain.

This finding raises the possibility that LBR arose in evolution by
the recombination of such an ancestral gene, encoding a polytopic
membrane protein, and another gene, encoding a soluble nuclear pro-
tein. LBR mRNA and protein was detected in several cell lines of dif-
ferent embryonal origin but at various amounts, suggesting regulated
expression of the LBR gene. The present results represent the first
determination of the structural organization of a gene that codes for an
integral membrane protein of the inner nuclear membrane.

99. The Polymorphic LY49 Family of Cell Surface Molecules Ex-
pressed by Natural Killer (NK) Cells. Hamish R. C. Smith, Brian F.
Daniels, Franz M. Karlhofer, and Wayne M. Yokoyama. Molecular
Medicine Division, Department of Medicine, Mount Sinai Medical Cen-
ter, New York, NY 10029 and VA Medical Center, San Francisco, CA
94121.*

We have recently demonstrated that the murine Ly-49 molecule is a
primary determinant of NK cell specificity. Ly-49 appears to interact
with polymorphic regions of major histocompatibility class I molecules
on target cells and globally inhibit NK cell activity. However, Ly-49 is

expressed only by a subset (20%) of NK cells. If the interaction between
Ly-49 and MHC class I represents a paradigm for understanding NK
cell specificity and putative inhibitory NK cell receptors, NK cells that
do not express Ly-49 should express other molecules, perhaps related to
Ly-49, that provide the inhibitory receptor function. To examine this
question, we used a full-length Ly-49 cDNA probe to screen a cDNA
library, prepared from oligo-dT-primed poly(A)* mRNA from IL-2-ac-
tivated NK cells.

Multiple clones were isolated and sequenced. All cDNAs distinct
from Ly-49 cDNA were analyzed in detail. If necessary, sequences were
extended by anchored-PCR of 5' regions.

These cDNAs were homologous to the Ly-49 cDNA with mis-
matches scattered throughout the sequences, indicating that they were
derived from distinct genes. The cDNAs were not allelic variants be-
cause all cDNAs were isolated from C57BL/6-derived cells. All cDNAs
predicted type II integral membrane proteins homologous to Ly-49.
Based on nucleotide homology, the cDNAs could be segregated into five
groups; one with Ly-49 cDNA itself (now termed Ly-49A) and four
other groups containing individual distinct sequences encoding poly-
peptides with 64-76% amino acid identity with Ly-49A. The final
group comprised 3 cDNAs that shared identical sequences but differed
from each other in coding or untranslated regions by deletions or in-
sertions of short DNA segments. By immunoprecipitation analysis
with an anti-peptide antiserum directed against a conserved portion of
the Ly-49A amino (intracellular) terminus, we demonstrated that Ly-
49" NK cells express Ly-49-related molecules.

Thus, NK cells express polymorphic molecules related to Ly-49A.
Further analysis of the expression and function of these molecules
should provide significant insight into NK cell specificity.

100. Regulation of Osteopontin (OPN) mRNA in Vascular Smooth
Muscle Cells (VSMC). Andrea S. Weintraub, Robert S. Green, Claire-
Lise Rosenfield, Mark B. Taubman. Mt. Sinai School of Medicine, Dept.
Pediatrics and Medicine, N.Y., N.Y.

Atherosclerotic coronary artery disease is characterized by increased
growth and migration of VSMC, resulting in intimal hyperplasia and
stenosis. Platelet-derived growth factor (PDGF), a potent mitogen and
chemoattractant for VSMC, is felt to play a role in the response of
VSMC to injury. To elucidate the mechanisms underlying PDGF-acti-
vation of VSMC, we have attempted to identify PDGF-inducible genes.

A rat aortic VSMC cDNA library was screened with cDNA from
quiescent (treated with 0.4% calf serum for 48 hr) and PDGF-stimu-
lated cells. Six clones were identified which demonstrated increased
hybridization to PDGF-stimulated cDNA.

The nucleotide sequence of one clone containing —1.5 kB insert
was homologous with rat OPN. OPN is a secreted protein involved in
matrix organization in bone and non-osteogenic tissues. Quiescent
VSMC expressed very low levels of OPN mRNA. 10% calf serum in-
duced high levels of OPN mRNA beginning at ~2 hr. PDGF (10 ng/ml),
angiotensin II (1 x 10 ""^ M), and basic fibroblastic growth factor (10
ng/ml) also produced a rapid induction of OPN mRNA; levels remained
elevated for 48 hr. TGF-p, which generally inhibits growth of cultured
VSMC, did not induce OPN mRNA.

These data suggest that stimulation of VSMC growth is accompa-
nied by increased OPN expression. The role of OPN in the vasculature
is unknown; its induction by growth agonists raises the possibility that
it may be involved in the regulation of VSMC growth, possibly by
providing the appropriate extracellular milieu.

101. Primary Structure of the Human Inner Nuclear Membrane
Lamin B Receptor and Interactions of the Expressed Protein with
Lamin B and DNA. Q. Ye and H. J. Worman. Division of Moleculsu-
Medicine, Dept. of Medicine and Brookdale Center for Molecular Biol-
ogy, Mount Sinai School of Medicine, New York, NY.

The lamin B receptor (LBR) is an integral membrane protein of the
inner nuclear membrane that was originally characterized in birds. It
has been proposed that LBR anchors the nuclear lamina to the inner
nuclear membrane by binding to lamin B in interphase. It has also
been hypothesized that LBR targets vesicular remnants of the inner
nuclear membrane to decondensing chromosomes at the end of mitosis
to initiate nuclear envelope reassembly in the daughter cells.

To further test these hypotheses and also establish if LBR is a
protein common to other vertebrate species, we have isolated cDNA
clones for human LBR, determined its primary structure and examined
the lamin B and DNA binding properties of expressed LBR polypep-
tides.

Human LBR is 68% identical to chicken LBR. It has a nucleoplas-

440

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

mie amino-terminal domain 210 amino acids in length followed by
eight putative transmembrane segments. The amino-terminal domain
is basic, has several DNA-binding motifs and contains consensus sites
for phosphorylation by protein kinase A and p34'^''''^ protein kinase
that are conserved in chickens and humans. A fusion protein contain-
ing the amino-terminal domain of human LBR can precipitate lamin B
from nuclear extracts of rat hepatocytes demonstrating that the nucle-
oplasmic domain binds to lamin B. This fusion protein also binds to
DNA on "Southwestern blots" and retards the migration of viral DNA
on agarose gels. Heat denaturation and protease treatment abolishes
the ability of the protein to bind to DNA. Like chicken LBR, human
LBR also undergoes mitosis-specific phosphorylation by p34'''^''^ protein
kinase that could hypothetically modulate lamin B and DNA binding
activities at different times in the cell cycle.

These data confirm that LBR is conserved among vertebrate spe-
cies. The nucleoplasmic amino-terminal domain of LBR can bind to
lamin B and DNA in vitro, consistent with LBR's proposed functions of
anchoring the lamina to the inner nuclear membrane in interphase
and in targeting inner nuclear membrane-derived vesicles to decon-
densing chromosomes during nuclear envelope reassembly at the end
of mitosis.

Renal

102. Cation Transport in MOCK Cells Transfected with Tamm-Hors-
fall Protein (THP). D. Kaji, J. Bates, J. Goyzueta, H. Yu, V. Sukhatme
and S. Kumar. Renal Division, Department of Medicine, VA Medical
Center, Bronx, NY, and Mount Sinai Medical Center, New York, NY;
University of Chicago, Chicago, IL; Beth Israel Hosp, Boston, MA; and
W. K. Warren Institute, University of Oklahoma, Oklahoma City, OK.

To examine whether expression of THP alters transport in renal epi-
thelial cells, we examined K influx in wild type MDCK cells (MD-WT),
MDCK cells transfected with THP using SV40 promoter (MD -i- 1), or
a cytomegalovirus (CMV) promoter (MD -i- 2) and MDCK cells trans-
fected with a control plasmid containing only the CMV promoter but no
THP (MD-1).

Expression of THP was confirmed by Northern blots and immuno-
cytochemistry. K influx, measured with ®^Rb in media containing (in
mM) NaCl 145, KCl 5, glucose 5, pH 7.4 at 37°C, with or without
inhibitors, was expressed as jimol • mg protein"' ■ hour"'. Means ±
SD of K influx, n = 6.

MD-WT MD -I- 1 MD 4-2 MD-1

Total 0.64 ± 0.04 0.29 ± 0.03 0.24 ± 0.03 0.29 ± 0.03
Ouabain

sensitive 0.34 ± 0.05 0.16 ± 0.06 0.11 ± 0.01 0.13 ± 0.03
Bumetanide

sensitive 0.23 ± 0.03 0.09 ± 0.04 0.09 ± 0.04 0.13 ± 0.02

Ouabain and
bumetanide

resistant 0.06 ± 0.04 0.03 ± 0.01 0.04 ± 0.01 0.03 ± 0.01

Ouabain and bumetanide resistant K influx was not different in all
groups. Ouabain sensitive and bumetanide components of K influx,
mediated by Na-K pump and NaK2Cl cotransport respectively, were
decreased in cells transfected with THP using SV40 or CMV promoter
(p < 0.005 compared to MD-WT). However, both ouabain and bumet-
Einide sensitive components of K influx were also reduced in MD-1, cells
transfected with the control plasmid containing only the CMV pro-
moter (p < 0.005).

Thus, cation transport in MDCK cells is unaffected by the expres-
sion of THP, but reduced by the process of transfection. The possibility
that the process of transfection itself may cause alterations in trans-
port should be considered in evaluating transport in transfected cell
lines.

103. Magnesium Depletion Increases NaK2Cl Cotransport in Medul-
lary Thick Ascending Limb (mTAL) Cells. D. Kaji, J. Goyzueta and J.
Diaz. Renal Division, Department of Medicine, VA Medical Center,
Bronx, NY and Mount Sinai School of Medicine, New York, NY.

The majority of salt reabsorption at the apical surface of mTAL is
mediated by the NaK2Cl cotransport. The mTAL is also an important
site for Mg reabsorption. While intracellular free Mg concentration is
known to modulate NaK2Cl cotransport in other tissues, the effect of
varying [Mg], on cation transport has not been evaluated in the mTAL.

We depleted [Mg], in cells cultured from mouse mTAL by incuba-
tion in low Mg media (Mg 0.1 vs 0.6 mM for control media) for 18 h.
(Mgli measured with FURAPTRA (Mg-Fura 2) was 0.55 ± 0.12 mM in
control cells and 0.27 ± 0.09 mM in Mg depleted cells (n = 8, p < 0.001
by paired t test). Activities of Na-K pump and NaK2Cl cotransport
were measured as ouabain sensitive and bumetanide sensitive (BS) K
influx respectively with *®Rb in media containing 145 mM NaCl, 5 mM
KCl, glucose 5 mM, pH 7.4 at 37°C, with and without ouabain (3 mM)
and bumetanide (10 jjlM).

Ouabain sensitive K influx was not diminished in Mg depleted
cells. However, BS K influx was 42% higher in Mg depleted cells com-
pared to control cells (10.82 ± 1.42 vs. 7.55 ± 1.19, n = 9, p < 0.001).
Affinity of NaK2Cl cotransporter for external Na, K and CI was not
different in Mg depleted cells. To examine whether the increase in BS
ion transport was associated with a similar increase in the number of
functioning cotransport sites, we measured ^[H] bumetanide binding.
^[H] bumetanide binding was not increased in Mg depleted cells com-
pared to controls (0.49 ± 0.11 vs. 0.62 ± 0.12 pmol ■ mg"').

We conclude that in contrast to erythrocytes, where Mg depletion
leads to a decrease in cotransport, Mg depletion in mTAL cells leads to
an increase in activity of NaK2Cl cotransport, as measured by BS K
influx. The increase in BS ion flux appears to be due to an increase in
turnover per transporter site, and not due to an increase in the number
of transporter sites.

104. Urea Acutely Inhibits NaK2CI Cotransport in Mouse Medullary
Thick Ascending Limb (mTAL) Cells. D. Kaji and J. C. Parker. Renal
Division, Department of Medicine, VA Medical Center, Bronx, NY and
Mount Sinai School of Medicine, New York, NY; and University of
North Carolina, Chapel Hill, NC.

Acute changes in medullary urea concentrations occur constantly with
the state of hydration and protein intake, but the effects of this change
on NaK2Cl cotransport have not been evaluated. Subdenaturing, phys-
iological urea concentrations may perturb protein structure, and result
in less macromolecular crowding, so that the cells behave as if they are
swollen, even at normal cell volume. Because cell swelling decreases
NaK2Cl cotransport in many cells, we examined whether urea had any
effect on NaK2Cl cotransport in mTAL cells.

NaK2Cl cotransport was measured in cultured mouse mTAL cells
as the fraction of K (*^Rb) influx that was sensitive to 10 \iM bumet-
anide.

NaK2Cl cotransport was markedly and progressively inhibited by
urea in a concentration dependent fashion, with 15% inhibition at 0.1
M urea and almost 80% inhibition with 0.5 M urea (p < 0.01 vs control
for all urea concentrations). These effects were completely reversible
and also observed with two other amides, acetamide and formamide.
Urea also inhibited NaK2Cl cotransport measured as CI dependent or
Na dependent K influx. Cell water content, measured with '''[C] 3-0-
D-methyl glucose was not different in urea treated cells compared to
control cells (3.61 ± 0.32 vs. 3.99 ± 0.32 jil/mg protein). Addition of 0.9
M-M okadaic acid or 10 nM calyculin, specific inhibitors of protein phos-
phatase 1 or 2A, prior to urea, blocked the inhibitory action of urea on
cotransport, suggesting that urea acts at a point before the phosphor-
ylation step.

Thus, increased urea concentrations may perturb protein struc-
ture, reduce the crowding effect of cytosolic proteins and by this mech-
anism cause a net dephosporylation of the cotransport protein or some
key regulatory protein that controls NaK2Cl cotransport. Acute in-
crease in urea concentration in the medulla may cause a m^or inhi-
bition of NaK2Cl cotransport in mTAL.

105. Classical and Channel-Like Urate Transporters in Rabbit Renal
Brush Border Membranes. Barbara A. Knorr, Jeanne C. Beck, and
Ruth G. Abramson. Division of Nephrology, Mount Sinai School of
Medicine, New York, New York.

The precise mechanism by which urate is transported across rabbit
renal proximal tubule luminal membranes has not been defined.

To determine whether urate flux across this membrane represents
simple diffusion or transport on specific carriers, urate uptake was
examined in brush border membrane vesicles that were prepared by a
magnesium aggregation technique and then exposed to CuCl2.

Urate uptake was consistently many fold in excess of chemical

Vol. 60 No. 5

ABSTRACTS

441

equilibrium. This concentrative, Na * independent uptake could not be
ascribed to non-specific binding or oxidation of transported urate to
allantoin. Uptake was voltage sensitive in that it was stimulated by
K^out + valinomycin (val), reduced in the presence of the permeant
anion thiocyanate, and reduced by the protonophore FCCP. The latter
suggests that uptake is driven, in part, by an H* diffusion potential
generated by copper's oxidation of SH groups. Consistent with this,
dithiothreitol (which prevents oxidation of SH groups) virtually oblit-
erated urate uptake. Uptake was cis inhibited by oxonate and trans
stimulated by urate. The cis inhibition was not due to dissipation of the
P.D. since oxonate did not alter the P.D. generated by K*out + val
(measured with the fluorescent, P.D. sensitive probe diS-C3-(5)). Trans
stimulation was evident even when the P.D. was short-circuited with
FCCP or "clamped" at a positive P.D. with K^o^t + val. Kinetic anal-
ysis using the 3-12 sec slopes of uptake revealed saturable and non-
saturable components. In paired studies oxonate (0.1 mM) inhibited the
non-saturable component: oxonate Eiffected the slope of uptake, not
binding.

These studies suggest that there are two mechanisms of urate
transport in rabbit brush-border membranes: (1) the demonstration of
voltage independent trans-stimulation and saturable uptake provides
strong evidence for a classical urate carrier; (2) the demonstration of a
non-saturable, but inhibitable component of urate transport indicates
that this portion of uptake is not simple diffusion but rather represents
facilitated diffusion on a potential sensitive channel-like carrier.

106. Expression of Urate-Anion Exchange in Xenopus laevis Oocytes.
Barbara A. Knorr, Jeanne C. Beck, and Ruth G. Abramson. Division of
Nephrology, Mount Sinai School of Medicine, New York, New York.

One mechanism of urate transport in rat renal proximal tubules is a
urate/anion antiporter that exchanges urate for a variety of organic
and inorganic ions.

As a first approach to cloning this anion exchanger we have em-
ployed the technique of expression cloning in Xenopus oocytes. Total
RNA was isolated by the method of Chirgwin et al. from rat renal
cortex. Poly A* RNA was selected by oligo-dt-cellulose chromatogra-
phy. Oocytes were injected with either 50 nl of total poly A* RNA (50
ng) or H2O. [2-^''C]urate uptake was measured on days 2-6 post injec-
tion in control and RNA injected oocytes incubated 60 min in ND96
media.

As maximal urate uptake/oocyte/hour was observed on day 3, all
subsequent data was obtained at that time point. Urate uptake was
stimulated 2-3 fold in RNA injected oocytes compared to those injected
with H2O. The stimulated uptake was completely abolished by the
disulfonic stilbene DIDS (1 mM), an anion exchange inhibitor. To eval-
uate the specificity of inhibition, uptake was also measured in the
presence of 1 mM oxonate. Oxonate, an inhibitor of a urate transporter
that has properties distinct from the anion exchanger, was without
effect. To determine the size of the RNA responsible for the stimulation
of urate upteike, total cortical poly A* RNA was fi-actionated on a
6-20% sucrose gradient. Individual fractions (5) were initially pooled
and urate uptake was compared in oocytes injected with 25 ng of each
pooled fraction. The 5 individual fractions from the pool that stimu-
lated urate uptake were subsequently injected in oocytes. Urate uptake
was maximally stimulated by the fraction containing 1.8 to 2.2 KB poly
A* RNA.

These studies indicate that the oocyte provides a model that per-
mits tremslation of the RNA of a rat urate trsinsport protein and its
insertion into the oocyte membrane. Since the resting membrane po-
tential of the oocyte ( - 55 mV) would oppose electrogenic urate uptake,
and transport is inhibited with DIDS but not oxonate, it is concluded
that stimulated transport must be due to translation of the electroneu-
tral urate/anion exchanger. These results indicate that it may be pos-
sible to use expression cloning to isolate the cDNA for the urate/anion
exchanger.

107. Isolation and Immunolocalization of a Uricase-Like Urate Trans-
port Protein. Barbara A. Knorr, Michael S. Lipkowitz, Barry J. Potter,
and Ruth G. Abramson. Divisions of Nephrology and Liver Diseases,
Mount Sinai School of Medicine, New York.

Two modalities of urate transport have been reported in rat kidney, a
urate/anion exchanger and a potential sensitive, uricase-like uni-
porter. As an initial attempt to isolate and characterize the responsible
transport protein(s), rat renal cortical membranes were harvested, sol-
ubilized, and subjected to affinity chromatography with urate or xan-
thine as the affinity ligand.

Bound protein was eluted with hypoxanthine. Pig liver peroxiso-

mal uricase was purified with the same system and the enzymatically
active purified protein was used to generate polyclonal antibodies in
rabbit.

In 6 of 10 studies the affinity purified renal protein oxidized urate
(following hypoxanthine removal): the specific activity was enriched
about 1000 fold compared to solubilized membranes. Silver stain of
SDS PAGE gels of the purified protein fraction revealed only 4 protein
bands at 27, 32, 36, and 41 kD. Western blot of the purified protein
fraction with anti-uricase demonstrated immunoreactive bands at 32
and 36 kD. Immunocytochemical studies of rat renal cortex with the
same antibody indicated that the immunoreactivity was localized to
the brush-border and subapical sites of proximal tubules. Finally, the
antibody to pig liver peroxisomal uricase produced a dose dependent
inhibition of urate uptake in renal cortical membrane vesicles (90% at
1 ug IgG/ug membrane protein), but did not alter Na-dependent glu-
cose transport.

These studies demonstrate that the renal cortical plasma mem-
branes contain urate binding proteins which have some functional and
immunological homology to the hepatic peroxisomal core protein,
uricase. Within the renal cortex, these proteins are localized to proxi-
mal tubules, the site of urate transport. Since the antibody which re-
acts with purified proteins on Western blot selectively inhibits urate
transport in intact membrane vesicles it is concluded that at least one
of the urate binding proteins is the uricase-like uniporter. It remains to
be established whether the 2 non-immunoreactive purified proteins
reflect epitopes of this transporter that are not recognized by the an-
tibody or monomers of a different protein, the urate/anion exchanger.

108. Reconstitution of Hepatic Uricase in Planar Lipid Bilayer Re-
veals a Functional Anion Channel. Edgar Leal-Pinto, Roger D. Lon-
don, Barbara A. Knorr, and Ruth G. Abramson. Division of Nephrolo-
gy, Mount Sinai School of Medicine.

Two mechanisms of urate transport have been detected in renal prox-
imal tubule cells: an electroneutral urate/anion exchanger and a urate
uniporter that is sensitive to the transmembrane electrical potential.
The protein responsible for renal membrane electrogenic urate trans-
port has a number of properties which indicates that it has some func-
tional and immunologic homology with uricase, a protein that resides
within the core of liver but not renal peroxisomes. Although uricase
functions as an oxidative enzyme in hepatocytes, we have demon-
strated that it can also function as a transport protein for urate when
inserted into the lipid bilayer of liposomes.

To further characterize the mechanism by which this protein
transports urate, uricase was reconstituted in phosphatidylcholine pro-
teoliposomes and then evaluated in planar lipid bilayers. Bilayers were
prepared with phosphatidylethanolamine, phosphatidylcholine and
cholesterol in a ratio of 16:3:1.

In the absence of uricase, no current was generated when the volt-
age was ramped between - 100 and +100 mV in symmetrical solutions
of 220 mM KCl, 2.5 mM urate, pH 7.4 on the cis and trans sides of the
bilayer. Insertion of uricase into the bilayer resulted in channel activ-
ity as evidenced by clear transitions between open and closed states.
The mean unitary slope conductance estimated by the analysis of sin-
gle channel records was 10 ± 1.0 pS. The current was voltage indepen-
dent and the channel displayed a high open probability (approximating
50%) with a mean open time > 100 msec. The channel did not conduct
K, Na, CI, or gluconate. Magnesium (0.25-1 mM) significantly de-
creased the open probability and the mean open time. Oxonate, a
known inhibitor of the enzymatic and transport activities of uricase,
produced a dose dependent inhibition (ion block) of urate transport on
the channel. In contrast to the effect of Mg, oxonate's inhibition was
characterized by a decrease in current carried by urate without a
change in open probability. In symmetrical 2.5 mM oxonate solutions
single channel conductance was at least 2 fold greater than in sym-
metrical urate solutions, indicating that the channel is more selective
to oxonate than urate.

These studies demonstrate that the non-membrane liver peroxiso-
mal protein uricase can behave as a highly selective anion channel
when inserted in a lipid bilayer. The ability of this protein to function
as a channel suggests the possibility that the renal proximal tubule
uricase-like protein transports urate electrogenically because it also
functions as an organic anion channel.

109. Regulation of Cystic Fibrosis Gene Expression in Renal Proximal
Tubule Cells by Parathyroid Hormone. Michael S. Lipkowitz. Nephrol-
ogy Division, Mount Sinai School of Medicine.

We have demonstrated that chronic changes in hormones that act via
the cAMP/protein kinase A messenger pathway modulate proximal

442

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

tubule CI permeability. In renal failure in rat, a state in which a
variety of hormones are chronically elevated, renal cortical homoge-
nate cAMP levels are elevated and accompanied by a selective increase
in brush border membrane CI permeability. Additionally, a chronic
infusion of parathyroid hormone, a hormone that is present in excess in
uremia and has been implicated as an important uremic toxin, also
results in a selective increase in CI permeability.

The cystic fibrosis gene product (CFTR) is a cAMP/protein kinase
A regulated chloride channel that is present in renal proximal tubule
brush border membranes. Studies were performed after a 3 day infu-
sion of parathyroid hormone to assess whether an alteration in the
expression (number) of CFTR channels is involved in the effects of
hyperparathyroidism on CI ~ permeability. Immunoblots using an an-
tibody to a peptide fragment of CFTR detect a 165 kD band consistent
with CFTR. This band is decreased by 25-30% in renal cortical brush
border membrane vesicles from hyperparathyroid compared to control
animals.

To begin to assess the mechanism by which CFTR protein expres-
sion is regulated, studies were performed to quantitate renal cortical
CFTR mRNA, Because of the paucity of CFTR mRNA in renal tissue,
reagents were generated to allow quantitation by competitive PCR.
PCR is performed simultaneously on sample and standard (CFTR DNA
with a new EcoRI restriction site) in the same tube; competition for
primers and reagents results in generation of products that reflect the
ratio of standard to sample. Subsequent digestion with EcoRI distin-
guishes sample (uncut, 270 bases) from standard (190 and 80 bases). In
studies employing first strand cDNA reverse transcribed from RNA
isolated from paired control and parathyroid hormone treated animals,
PTH induced a 23% decrease in CFTR mRNA.

Previous studies have demonstrated regulation of CFTR expres-
sion in colon carcinoma cell lines; the current studies are the first data
that suggest that cystic fibrosis gene expression can be regulated by
physiologic stimuli in vivo. Since the decrement in CFTR mRNA is
approximately the same as that in CFTR protein, these data suggest
that CFTR expression may be regulated at the level of transcription by
parathyroid hormone. This decrease in expression (number) of CFTR
CI " channels may be a counter-regulatory response aimed at preserv-
ing cell membrane potential and intracellular electrolyte concentra-
tions in the face of the increased CI" permeability (channel activity)
induced by PTH.

110. Regulation of Red Blood Cell Chloride Permeability Is Dependent
upon Normal Function of the Cystic Fibrosis Gene Product CFTR.

Michael S. Lipkowitz and Roger D. London. Division of Nephrology,
Mount Sinai School of Medicine, NY.

We have previously demonstrated that human red blood cell mem-
brane chloride permeability is hormonally regulated via the adenylate
cyclase/cAMP/protein kinase A messenger pathway. Increases in red
cell cAMP induced by epinephrine and norepinephrine resulted in a
significant increase in the permeability of chloride relative to that of
potassium (PCl/PK). Direct activation of red blood cell adenylate cy-
clase by forskolin markedly increased intracellular cAMP concentra-
tion, and also significantly increased PCl/PK. Finally, in renal failure,
a pathological state in which norepinephrine and epinephrine levels
are elevated, red blood cell cAMP and PCl/PK are elevated; both cAMP
concentration and PCl/PK are normalized by correction of renal failure
with dialysis.

Since the cystic fibrosis gene product (CFTR) is a PKA activated
chloride channel (or chloride channel regulator) that is present in
many tissues, studies were performed to determine whether this pro-
tein could be responsible for the regulation of RBC chloride permeabil-
ity.

Red blood cell membranes were solubilized in SDS, de-lipidated by
methanol/chloroform extraction, and separated by SDS-PAGE. Immu-
noblots using a rabbit polyclonal primary anti-CFTR antibody with
detection by a peroxidase-linked secondary antibody and enhanced
chemiluminescence demonstrated the presence of a 165 kD band con-
sistent with CFTR.

Red blood cells from patients with cystic fibrosis were incubated
for 30 min with or without 10 uM forskolin, after which cAMP was
quantitated with an isotopic technique and PCl/PK was determined
with the fluorescent, potential sensitive dye diS-C3-(5). Forskolin pro-
duced a marked increase in red cell cAMP (from 44.5 ± 20 to 160 ± 38
pmol/ml cells). Despite this increase in cAMP, forskolin did not produce
a significant increment from baseline in PCl/PK (10 ± 6%) in cystic

fibrosis patients as opposed to a clear increment in PCl/PK in controls

(94 ± 6%).

These studies suggest that the cAMP induced increase in red blood
cell membrane PCl/PK is dependent on normal function of the cystic
fibrosis gene product, CFTR. Regulation of chloride channel activity
has been demonstrated to play an important role in maintenance of cell
volume and in the normal function of intracellular organelles in a
variety of tissues. The significance of regulation of chloride channels in
mature red blood cells remains to be clarified; however, such regulation
may play an important role in maintaining homeostasis in the more
metabolically active red cell precursors.

111. Chronic Renal Failure Reduces P-Glycoprotein Expression in
Rat Brush Border Membranes. Roger D. London, Edgar Leal-Pinto,
and Shih Chang. Division of Nephrology, Mount Sinai School of Med-
icine, New York.

The human multidrug resistance (MDR) gene product, P-glycoprotein
actively transports cytotoxic drugs out of cells. Cytotoxic drugs and
acute environmental stresses such as heat shock and arsenite can in-
duce P-glycoprotein expression. (In addition expression of P-glycopro-
tein has recently been associated with a cell volume-regulated chloride
channel.) In the present study, the expression of P-glycoprotein was
examined in a rat renal ablation model of chronic renal failure.

Kidneys were harvested from control, uni-nephrectomized and
75% nephrectomized rats that were pair-fed for 2-3 months after sur-
gery. Renal cortical brush border membrane vesicles were prepared by
isolation on a 15% percoll gradient. P-glycoprotein expression was
evaluated by Western blots of brush border membranes and cortical
homogenates using C219 monoclonal antibody. Levels of P-glycopro-
tein expression were quantitated using laser densitometry. Expression
of villin, a marker of brush border membrane was also evaluated.

We have previously determined that 75% nephrectomy reduces
glomerular filtration rate to approximately one third of control values.
In five groups of control and renal failure animals there was a signif-
icant reduction (46%, p < 0.03) in the expression of P-glycoprotein in
brush border of renal failure animals, while the amount of P-glycopro-
tein in the control and renal failure cortical homogenates was not sig-
nificantly changed. In contrast, expression of villin was not altered by
chronic renal failure. P-glycoprotein and villin expression were not
altered by the hypertrophy associated with uni-nephrectomy.

In view of the ability of certain substrates to inhibit their own
transport on P-glycoprotein, it is suggested that in chronic renal failure
the reduced amount of P-glycoprotein in renal brush border mem-
branes detected in the present study would further reduce the excretion
of cytotoxic drugs and/or cyclosporine. Thus reduced renal transport
and systemic accumulation of the substrates for P-glycoprotein could
sjTiergistically enhance their toxicity in chronic renal failure.

112. Ion Channels in Renal Failure. Roger D. London, Michael S. Lip-
kowitz, and Edgar Leal-Pinto. Division of Nephrology, Mount Sinai
School of Med., New York.

Studies from our laboratory have reported an increase in the ionic
permeability of chloride relative to that of potassium (PCl/PK) in red
blood cell ghosts of uremic humans and in renal cortical brush border
membranes and RBC membranes prepared from rats in renal failure.
These observations suggest that membrane ionic permeability is per-
turbed in renal failure in both polarized and non-polarized cells. In
both human and rat membranes, the alteration in membrsme perme-
ability was consistently associated with an increase in intracellular
concentrations of cAMP in red blood cells and renal cortical homoge-
nates. Since the level of a variety of hormones (PTH, norepinephrine,
glucagon, and ADH) that act via the cAMP pathway are increased in
renal failure, and cAMP, PKA, PKC and G proteins have been impli-
cated in regulating ion channels, it seemed possible that a cause and
effect relationship might exist between the increased PCl/PK and
[cAMP] in renal failure. To evaluate whether the open probability
and/or number of chloride channels was increased in chronic renal
failure, we directly characterized the individual ion channels and ex-
amined their regulation.

In preliminary studies, rat BBMV isolated on a 15% percoll gra-
dient were reconstituted in planar lipid bilayers. In symmetrical 150
mM Cesium or TEA chloride, small 7-10 pS channels were observed.
These channels were non-rectifying, selective for chloride over cations
and inhibited by the chloride channel blocker DPC. Furthermore, ad-
dition of protein kinase A (150 units) with 1 mm ATP increased the
open probability of channels from normal rat BBMV. Of note, two
chloride channel proteins, P-glycoprotein and the cystic fibrosis trans-

Vol. 60 No. 5

ABSTRACTS

443

membrane conductance regulator (CFTR) have been localized to the
BBMV of renal proximal tubules, and there is evidence that their reg-
ulation is, at least in part, due to protein kinase A phosphorylation.

Thus, further studies were performed to evaluate whether the in-
crease in relative ionic chloride permeability is secondary to an in-
creased number of chloride channels in renal failure. When P-glyco-
protein and CFTR expression were evaluated by Western blotting,
levels of both P-glycoprotein and CFTR were decreased in renal failure
brush border membranes, while villin, a marker of brush border mem-
brane was not changed. There was no detectable change in renal cor-
tical homogenates for P-glycoprotein, CFTR or villin. These studies
suggest that the amount of two chloride channel proteins is decreased
in renal failure. Thus, if the increased PCl/PK is a result of an increase
in chloride conductance, and not a fall in potassium conductance, the
increased chloride current would be mediated by fewer channels with
an increased open probability. Since hormonally induced chronic ele-
vation of [cAMPl would phosphorylate chloride channels increasing
their open probability, an adaptive decrease in channel number would
act to restore membrane potential towards normal and preserve the
transport functions of the renal brush border.

Oncology /Neoplastics

113. Human Tumor Sensitivity Testing Finds New Treatment Options.

H. W. Bruckner,^ P. A. Andreotti,^ I. A. Cree,^ K. Becker," P. A.
Caruso^ and I. Gleiberman.^ 'Mount Sinai School of Medicine, N.Y.;
'^Batle LE Division, Ft. Lauderdale, Fl; ^Dundee University, Dundee,
Scotland; and "Eppendorf University, Hamburg, Germany.

An ATP bioluminescence chemosensitivity assay system evaluated dis-
sociated tumor cells cultured for 5-7 days with proprietary medium in
microplates. The assay tests 8 drugs or combinations at 7 concentra-
tions in triplicate using as few as 20,000-40,000 cells obtained by
surgery or needle biopsy. The system finds heterogeneity of chemosen-
sitivity between tumors as well as additive or synergistic drug effects.
It provides supporting insight into some options for current treatment
of ovarian cancer (cisplatin-based), breast cancer (adriamycin-based),
and AML (mitoxantrone-based).

Initial results with ovarian, breast, and AML specimens indicate a
predictive accuracy of >70% for sensitivity and >95'^ for resistance.
Studies with 300 specimens have demonstrated the important role of
correlations with clinical experience in order to define optimum in vitro
assay conditions and criteria for definitions of sensitivity and resis-
tance for each class of tumor.

The system appears to have application for both drug development
and selection of regional and combination therapy. The in vitro expe-
rience with mesothelioma human cell lines carried in nude mice con-
firmed all known in vivo sensitivities and resistances as well as the
relative effectiveness of treatment options. Taxol was identified as a
promising new drug for mesothelioma. Taxol-Adriamycin combina-
tions were confirmed to be particularly active and an encouraging clin-
ical lead for investigation of both ovarian cancer and mesothelioma.

114. Polypeptide Markers in Pancreatic Cancer. H. W. Bruckner,
M. R. Chesser, J. DiGiovanni, L. Farber, J. Mandeli. The Derald Rut-
tenberg Cancer Center & Department Biomathematics, The Mount
Sinai Medical Center, N.Y.

Patients with duct cell pancreatic cancers have greater than normal
concentrations of polypeptides (gastrin, substance P, chromogranin,
pancreatic polypeptide neurotensin, and gastrin release peptide) in
their peripheral blood. (Bruckner et al. AACR 92,93). One-third of
patients have a peptide in abnormally high concentrations. GRP is
present in 50% of patients with stage IV disease but only <5% of those
with regional disease. Substance P appears to have an association with
superior survival, the median approaching 28 and 18 months in pa-
tients with stage 111 and IV disease respectively. Although gastrin is a
growth factor, unexpectedly it continues to have an association with
superior survival, the median approaching 24 months and 13 months
in patients with stage III and IV disease respectively. Chromogranin
and pancreatic polypeptide are not associated with good survival. One
sixth had 2 peptides in abnormally high concentrations.

Polypeptide production may represent experiments of nature of-
fering insight into physiologic control of pancreatic tumors. Their pres-

ence in abnormal concentration is not conclusive evidence that the
physician has discovered an endocrine tumor.

115. Triple Biochemical Modulation for Metastatic Adenocarcinoma
of the Stomach. H. W. Bruckner, J. Morris, A. Jennis, A. M. Gattani,
J. Mandeli, D. Lehrer and M. R. Chesser. Mount Sinai School of Med-
icine, The Derald Ruttenberg Cancer Center, N.Y.

This regimen, MLP-F, utilizes methotrexate (M), leucovorin (L), and
cisplatin (P) in order to modulate bolus and infusional fluorouracil (F).
The MLP-F regimen has been described as producing high rates of
response and median survival in excess of 1 year for patients (pts) with
metastatic adenocarcinomas of the stomach and pancreas (Bruckner et
al., Clin Gastroenterol, 1991; and Gattani et al., ASCO, 1992).

Forty-one consecutive eligible pts with adenocarcinoma of the
stomach and lower esophagus were given MLP-F. Patient characteris-
tics: All pts had either unresectable disease or residual disease follow-
ing palliative surgery. All pts had a performance status of 0-2. Median
age was 60 years, but 13 pts were less than 50 years old. Thirty pts
were males. Palliative resection was performed for 15; 17 had no sur-
gery; and 9 were explored only. Thirty-four pts were given MLP-F as
initial treatment.

Overall median survival for this treatment group was 15.8
months. Overall response rate was 59%. The complete response rate
was 27% . Thirty percent of all pts survived more than 2 years. Three of
the 34 pts discontinued MLP-F electively and subsequently progressed,
and 7 additional pts were treated with MLP-F after receiving other
prior chemotherapy. The response rate in these 10 pts who had received
prior chemotherapy was 60%. Median survival from start of MLP-F for
pts with prior chemotherapy was 9.7 months.

Unlike experience with other chemotherapy, peritoneal disease
does not override the impact of MLP-F therapy. When the patient's
survival was examined, based on a published survival model which
employed nine objective prognostic laboratory tests (such as albumin,
bilirubin, CEA, white blood count), even the pts with poor prognostic
features achieved a median survival approximating 1 year; this sur-
vival is better than that of the historical group with good prognostic
characteristics.

116. Tumor Markers Provide Objective Prognostic Information in the
Context of Combined Modality Pancreatic Cancer Trials. A. M. Gat-
tani, J. Mandeli, and H. W. Bruckner. Mount Sinai School of Medicine,
The Derald Ruttenberg Cancer Center, New York, N.Y.

Patients with stage II -III pancreatic cancer (PC) were treated with
regimens using combined modality methods (Bruckner et al., AACR
'89 and Kamthan et al., ASCO '92) and biochemical modulation (Gat-
tani et al., ASCO '92; Bruckner et al., J Clin Gastroenterol '91 and
Cancer Res '88). Pretreatment tumor markers, carcinoembryonic anti-
gen (CEA), CA 19-9, and CA-125 (Dianon Laboratories) were evaluated
regarding their correlation with survival in 34 patients.

Analyses revealed naturally occurring gaps in the distribution of
assay values. This gap divided patients into "good" vs. "bad" risk
groups. A CA 19-9 assay of 2,000 Wml divided patients into two groups;
23 'low assay' patients (<2,000) had a median survival time (MST) of
17.9 months (mos), 11 'high assay' patients (5=2,000) had a 7.4 mos
MST, (p.0002). A CEA assay of <10 mg/ml identified 28 patients as a
low assay group, 17.9 mos MST, and 6 patients as a high assay (slO
mg/ml) group with MST of 6.1 mos (p. 001). Simultaneous assays for
example, CA 19-9 plus CEA low assays found 19 "good" prognosis pa-
tients with a MST of 22.8 mos (p.0002). CA 19-9 and CA 125 high
assays found 6 "poor" prognosis patients with MST of 6.8 mos (p. 0001).

Conclusions. The assays assigned the majority of patients to good
risk groups. The MST exceeded that found using any published prog-
nostic test for similar unresectable stage II-III PC. The "bad" risk
group has an equal or better survival compared to historical patients
overall. These methods of prognostication contradict the normal expec-
tations for a majority of patients and identify a high priority group for
treatment. These results warrant the use of baseline tumor markers in
the design (stratification) and analysis of clinical trials for the treat-
ment of pancreatic cancer. That the majority of patients achieve a
survival of heretofore unexpected length implies that the combined
modality treatment may contribute favorably to the survival of pa-
tients.

117. High Dose Zidovudine (AZT) in Patients with Treated Myelodys-
plastic Syndromes (MDS), and Relapsed or Refractory Acute Leuke-
mia. L. R. Silverman, J. Roboz, E. P. Demakos and J. F. Holland. The
Derald H. Ruttenberg Cancer Center and The Department of Neoplas-
tic Diseases, Mount Sinai School of Medicine, New York, NY 10029.

444

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

AZT, a chain terminator of DNA synthesis with myelosuppressive ef-
fects in patients with AIDS, can inhibit the growth and induce the
differentiation of HL-60 leukemia cells in vitro.

AZT was administered po q6h in divided doses rounded off to the
nearest 100 mg, according to diagnosis and preset cohorts (MSSM GCO
#89-306 NPD*): 900 mg/m^/d (n = 3); 1100 (n = 5); 1300 (n = 1); 1500
(n = 4); 1700 (n = 3). Doses were escalated in 3 patients. Ten of 11
patients diagnosed with MDS (2), transformed (tx) AML following
MDS (2). tx biphenotypic leukemia following MDS (3), refractory AML
(1) relapsed biphenotypic AML (1) and secondary AML (2) were eval-
uable. All patients had failed at least one prior treatment regimen
(median 2). The median age was 63 (22-76); 5 were male and 5 had a
PS 0-1.

Responses occurred in 5; 1 CR (MDS); 2 PR (tx AML, biphenotypic
AML); 2 Improved (2 tx AML). Marrow aplasia did not occur in any
patient. Grade 4 leucopenia occurred in one patient and thrombocyto-
penia occurred in three. Increased skin pigmentation was seen in four
patients. Pharmacokinetic studies by HPLC demonstrated peak AZT
serum levels between 4.3-16.4 \iM and AZT glucuronide peak levels
between 8.6-48.8 jjlM.

No correlation of dose and hematologic effect has yet been estab-
lished. Peak levels (30'-90') tended to increase within single patients
as doses escalated. Absolute doses of AZT ranged from 1200-3600
mg/day up to 7 x higher than the dose in AIDS. High dose AZT was well
tolerated and produced clinical responses, without bone marrow apla-
sia, in a heavily pretreated group of patients with advanced myelodys-
plasia and derivative leukemias.

Pulmonary

118. Pyrazinamide-Induced Hepatic Necrosis Treated with Liver
Transplant. J. J. Adler, B. Meyers, C. Miller, F. Klion. Departments of
Medicine and Surgery, Mount Sinai Medical Center, N.Y.

While pyrazinamide (PZA) is a recognized cause of drug-induced hep-
atitis, hepatic necrosis remains a rare complication of this drug. In the
early years of TB chemotherapy PZA was relegated to "second line"
status in part because of its hepatic toxicity when compared to other
drugs. In recent years PZA has been used as "first line" treatment
because of its specific tuberculocidal effect on intracellular organisms
and its efficacy in short course treatment regimens. With the increased
fi-equency of multiple-drug-resistant tuberculosis (MDR TB), greater
use of PZA for both the treatment of active disease and preventive
therapy in individuals thought to be infected with MDR TB is antici-
pated.

A 30-year-old black woman whose husband had documented MDR
TB with an organism resistant to isoniazid (INH), rifampin, strepto-
mycin, and ethambutol, along with her 2-year-old daughter converted
their tuberculin skin tests. Initially, the patient took INH preventive
therapy, but when the drug resistance pattern became known, she was
placed on PZA (20 mg/kg) and ofloxacin (800 mg) and INH was discon-
tinued. Laboratory studies showed elevation of ALT and GOT 3 times
baseline prior to the discontinuation of INH. Three weeks after start-
ing PZA the patient noted nausea, vomiting, scleral icterus, and dark
urine; she stopped all medication immediately. However, over the next
ten days she developed progressive hepatic failure, necessitating emer-
gency liver transplant.

This person in close contact with a patient with MDR TB is highly
instructive in pointing out the risk of hepatic necrosis with indicated
therapy for MDR TB. In addition, with the need for prolonged immu-
nosuppressive treatment, the scope of tuberculosis prevention in liver
transplant patients with both sensitive and multiple resistant organ-
isms will increase.

119. Transbronchial Biopsy (TBB) Significantly Enhances Yield of
Bronchoscopy in HIV-infected Patients; 1992. L. Bernstein, S. Salz-
man, P. Villamena, M. Rosen. Division of Pulmonary and Critical Care
Medicine, Beth Israel Medical Center (BIMC), N.Y.

Fiberoptic bronchoscopy (FOB) techniques used to diagnose HlV-re-
lated pulmonary disorders vary by institution. The high frequency of P.
carinii pneumonia (PCP) and the high sensitivity of bronchoalveolar

lavage (BAD in diagnosing PCP has led to BAL use alone during FOB.
PCP prophylaxis and sputum induction have reduced the number of
cases requiring FOB and may alter FOB yield.

To evaluate TBB and BAL yield, we reviewed the records of all
patients who had FOB at BIMC in 1992. Of 519 FOBs performed, 205
were in 182 known or suspected HIV-infected patients.

37 cases were nondiagnostic, 180 diagnoses were established and
13 patients had dual diagnoses as shown below:

Diagnosis N (%) BAL smear* TBB path* TBB -I- /BAL -t

PCP

105

(58)

90/105

(86%)

85/99

(86)

15/106

(14)

KS (13/15 visual)

(8)

4/10

(40)

2/15

(13)

Bacterial pneum.

(8)

4/15

(27)

4/15

(27)

CMV

(7)

4/12

(33)

11/12

(92)

8/12

(67)

Inter. pneum/Fib.

(7)

0/12

(0)

12/12

(100)

12/12

(100)

M. tuberculosis

(3)

6/6

(100)

1/5

(20)

0/6

(0)

M. avium comp.

(2)

2/3

(67)

0/3

(0)

0/3

(0)

Fungi

(3)

3/5

(60)

3/4

(75)

1/5

(20)

Non-KS malign.

(3)

3/5

(60)

4/4

(100)

2/5

(40)

Nocardia: 1 Cx-i-

(1)

1/2

(50)

0/2

(0)

0/2

(0)

Total diagnoses

180

(100)

109/150

(73%)

124/166

(75%)

44/180

(24%)

* Specimens positive obtained

+ Specimens positive/total diagnosed.

Conclusions. PCP remains the most common disease diagnosed by
FOB. Diagnoses were made exclusively by TBB in 24% of cases includ-
ing 14% of PCP cases. Some disorders resembled PCP clinically, but
were only diagnosed by TBB. When performing FOB in HIV-infected
patients, TBB significantly enhances diagnostic yield over BAL alone.

120. Connexin43 Expression Regulates Growth in Cultured Human
Airway Smooth Muscle Cells (HASM): 1992-93. Louis R. DePalo. Di-
vision of Pulmonary Medicine, Mount Sinai Hospital, NY.

Direct intercellular communication via gap junctions is an important
mechanism allowing the coordination of tissue function on a cellular
level. Gap junctions are formed by the apposition of cell membreines
containing hexameric assemblages of structural proteins termed con-
nexins. The intercellular exchange of second messenger molecules and
ions between coupled cells may amplify the effects of localized nervous
or humoral stimulation on synchronized tissues, and perhaps synchro-
nize metabolic as well as electrical activity in the tissue. This exchange
of cellular information is used to coordinate diverse cellular functions
such as pattern formation in development, electrotonic coupling in the
nervous system and smooth muscle viscera and specific tissue re-
sponses to injury (liver). Recently, it has been shown that the lack of
formation of functional gap junctions in human carcinoma can be cor-
related to metastatic potential, suggesting that the gap junction may
be an important regulator of cellular growth and contact inhibition.

We have previously shown that cultured tracheal myocytes ex-
press a human homologue of rat connexin43, the major gap junctional
protein in heart and vascular tissues. Expression of connexin43 in this
tissue results in dye- and electronic coupling with a mean junctional
conductance of 9.43 ± 1.01 nS (SEM, N = 41) and unitary conductance
of about 90 pS. Inhibition of connexin43 expression by antisense oli-
gonucleotides (Cx-AS) significantly inhibited both dye coupling and
levels of connexin43 detectable by immunoblot (0.59% ± 0.09 SD).
Junctional conductance (gj) was greatly inhibited by Cx-AS [(17.75 ±
3.0 (SE), N = 4) vs. 4.39 ± 3.06, N = 9) (P s .05, Students t-test)].

HASM cells grow to confluence in response to PDGF and demon-
strate significant contact inhibition with a flattening of the growth
curve and marked diminution in bromodeoxyuridine incorporation.
Cell cycling at confluence constitutes approximately 1-4% of the cell
population. Addition of 100 jjiM Cx-AS resulted in a significant in-
crease in cell number of approximately 0.20% ± 0.005 (N = 4) in
response to serum and evidence of an increase in S phase cycling as
determined by incorporation of BUDR over controls treated with a
nonsense oligonucleotide. In contrast, Cx-AS addition to confluent rat
aortic smooth muscle (RASM) cells (expressing connexin43 without
significant dye coupling or significant contact inhibition) showed no
difference in cell counts compared to non-sense treated controls.

Cultured HASM express connexin43 and form functional gap junc-
tions. Connexin43 expression and functional gap junction formation
can be significantly inhibited by the addition of antisense oligonucle-
otides to the connexin43 gene. Inhibition of gap junction formation in
confluent cultures results in loss of contact inhibition with resultsint

Vol. 60 No. 5

ABSTRACTS

445

cell cycle entry and cellular proliferation. Smooth muscle hyperplasia
in airway disease may thus be regulated in part by gap junctions. Cx43
antisense inhibition, however, fails to result in proliferation of RASM,
cells that express connexin43 without tight functional coupling and
that lack significant contact inhibition. The precise role of gap junction
formation and intercellular coupling in regulating proliferation in dif-
ferent tissues deserves further study.

121. The Effect of High Dose Steroids on Sleep Patterns and Respira-
tory Function in Symptomatic Steroid-Dependent Asthmatics. Fein E,
Godbold J, Chusid E, Schachter EN. Mount Sinai School of Medicine
New York, NY 10029.

Nocturnal and early morning wheezing is common among asthmatics,
yet the mechanism and therapeutic approach are unclear. Some studies
indicate a correlation between the severity of asthma to these sjonp-
toms and the frequency of nocturnal awakenings.

We recruited ten adult steroid dependent asthmatics on a stable
(at least three months) dose of the equivalent of prednisone &7.5 mg
but e20 mg. Following Albuterol, all subjects demonstrated improve-
ment of lung function of at least 15% of the FEVi. All patients signed
informed consent approved by our Institutional Review Board. Each
patient was studied for a four week period. During the first two weeks,
the patients were followed on their standard medications. During the
second two week period, patients were placed on 40 mg of oral pred-
nisone while continuing their usual bronchodilator therapy. Subjects
were provided with a questionnaire and were asked to record the num-
ber of times they were awakened due to their asthma during the pre-
vious night. Additionally, they reported the severity of symptoms
(scale 0-3) for wheeze, dyspnea, £ind cough and overall symptoms
(scale 0-4). They also recorded the number of puffs of bronchodilator
and theophylline tablets used during the previous twenty-four hour
period. Finally peak flow rates were measured in the morning and in
the evening. There were 6 female and 4 male asthmatics studied, aged
22-66.

The baseline lung function of our subjects was: FVC 90% ± 13%,
and FEVi 63% ± 12% expressed as a percent of predicted. The mean
dose of prednisone on entry to our study was 12.5 mg. ± 3.35. The mean
number of awakenings during the second week of the study was 0.7 ±
0.3, and 0.20 ± 0.05 during the fourth week of the study (p = 0.0001).
A trend of increasing AM and PM peak flows were noted for the group
during weeks three and four. Whereas the mean scores for wheeze,
dyspnea, overall symptom score, puffs of bronchodilator, cough, and
number of theophylline tablets per day improved significantly when
comparing week two to four.

We conclude that stable symptomatic steroid dependent asthmat-
ics suffer frequent nocturnal awakenings. Increased corticosteroid dos-
ages in such individuals rapidly decrease the number of awakenings
and concurrently improve other respiratory symptoms.

Supported in part by Marion Merrell Dow, Inc.

122. Mucin Gene Expression in an Endometrial Adenocarcinoma Cell
Line. E. G. Gollub, S. Goswami, D. Kouba and Z. Marom (SPON: E.
Schwartz). Division of Pulmonary and Critical Care Medicine, Mount
Sinai Med. Center, New York, N.Y. 10029.

The endometrial adenocarcinoma cell-line (Ishikawa cells) secretes a
mucin which resembles airways epithelial mucin. Cells treated with
nanomoleu- amounts of calcium ionophore, A23187, exhibit greatly en-
hanced mucin secretion.

To examine the regulation of mucin secretion at the molecular
level, cells were treated with A23187 and the expression of mucin
genes was analyzed by Northern blot hybridization. Five distinct mu-
cin genes have been identified (MUC 1-MUC 5) to date. Synthetic
oligonucleotides corresponding to the sequences of MUC 1-MUC 5
were used as hybridization probes.

Only MUC-4, originally isolated from a tracheobronchial cDNA
library, hybridized to RNA prepared from control and A23187-treated
cells, exhibiting a discrete band slightly below the 28S RNA (~5 kb).
The RNA from the stimulated cells gave a 3-4 x higher signal. Treat-
ment of the cells with other secretagogues, hydroxyeicosatetraenoic
acid (HETE) and carbachol) also up-regulated the expression of MUC 4.

Thus, enhanced mucin secretion in Ishikawa cells is accompanied
by increased levels of MUC-4 mRNA. These results may indicate an
effect on (a) RNA transcription, or stability, or (b) in the case of the
ionophore, a general effect on mRNA levels related to an increased
influx of calcium.

Aided by a gremt from the New York Lung Association.

123. Phosphoinositide Hydrolysis and Protein Kinase C Activation by
15 Hydroxy Eicosatetraenoic Acid Induces Mucin Secretion in an E)p-
ithelial Cell Line. S. K. Goswami and Z. Marom. Division of Pulmo-
nary and Critical Care Medicine, Mt. Sinai Hospital, New York.

We have demonstrated previously that 15-hydroxy eicosatetraenoic
acid (15-HETE) is a potent mucus secretagogue in human airway ex-
plants in culture. We have established an epithelial cell line (Ishikawa
cells) which secreted mucin like glycoprotein (MLGP) under a variety
of external chemical stimuli.

Since 15-HETE enhances mucin secretion in human airway ex-
plants, the present investigation was undertaken to study the effect of
15-HETE on MLGP secretion, if any, in Ishikawa cells and the under-
lying mechanism(s) involved therein. 15-HETE (2.5 ji.g-10 jjig/ml) in-
duced a dose-dependent increase of MLGP (20 ± 2% to 60 ± 2% in-
crease over control, n = 4). It has been demonstrated by us that rise in
intracellular Ca * * is vital for the secretion of MLGP in these cells.
Intracellular Ca * * is altered by the generation of inositol triphos-
phate during phosphoinositide (PI) hydrolysis. We, therefore, studied
the effect of 15-HETE on PI hydrolysis in [3H] myo-inositol labeled
cells.

PI turnover was increased by 15-HETE in a dose-dependent and
time dependent fashion. Protein Kinase C (PKC) activation in these
cells by phorbol myristate acetate (PMA) has been shown to enhance
MLGP secretion. It is feasible that 15-HETE by generating diacylglyc-
erol during PI hydrolysis could activate PKC and induce MLGP secre-
tion. In order to substantiate this hypothesis, we used (i) a PKC inhib-
itor, l-(5 isoquinolinyl sulfonyl)-2-methyl piperazine (H-7), and (ii)
down-regulation of PKC by incubating the cells with PMA for 6 hours.
In both instances the MLGP secretion by 15 HETE was inhibited.

These findings strongly suggest that 15-HETE exerts its mucus
secretagogue activity by the activation of PKC.

Supported by NIH Grant 37254.

124. Phospholipase C Activity in Human Respiratory Smooth Muscle
Cell Is Inhibited by Both Protein Kinase A and Protein Kinase C
Medicated Phosphorylation. S. K. Goswami and Z. Marom. Division of
Pulmonary and Critical Care Medicine, Mount Sinai Hospital, New
York.

Human respiratory smooth muscle cell (HRSMC) was isolated from
human airways by established methods. Immunohistochemical stain-
ing of these cells showed the presence of myosin, smooth muscle actin
and desmin. Actin and myosin's smooth muscle origin was further
identified by the presence of m-RNA encoding a-actin and smooth mus-
cle myosin light chain by blot hybridization with ^^P labeled CDNA
probe. Bradykinin (BK) and sodium fluoride (NaF) were used as acti-
vators of phospholipase C (PL-C). PL-C activation induces polyphos-
phoinositide (PI) hydrolysis and generation of myo-inositol phosphates
(IPs). [3H] Myo-inositol labeled HRSMC was used to measure IP's pro-
duction by BK and NaF.

BK (10 ''M to 10 *M) enhances PI hydrolysis in a dose-dependent
manner. NaF induces the release of IP's in a linear fashion with opti-
mal induction occurring at 25 mM. Both effects were blocked (about
50%) by pre-treatment of cells with forskolin (10 jjiM), a Protein Kinase
A (PKA) activator. Phorbolmyristate acetate (PMA), an activator of
Protein Kinase C (PKC), also inhibited PI hydrolysis induced by BK
and NaF.

Since the site of activation by NaF is believed to be a GTP binding
protein and forskolin inhibits both BK and NaF induced PI hydrolysis
it is very tempting to speculate that a GTP binding protein is involved
in coupling the BK receptor to PL-C. Forskolin, by activating PKA,
could mediate the phosphorylation of this GTP binding protein and
inhibit PL-C activity.

Supported by NIH Grant 37254.

125. Isolation and Characterization of a Novel Epithelial Cell-Specific
T Lymphocyte Ligand. TH Kalb, M Walter, Y Li, and L Mayer. Divi-
sions of Pulmonary/Critical Care and Clinical Immunology, Mount
Sinai Hospital, NY.

Introduction. Airway (AEC) and intestinal (lEC) epithelial cells have
been previously shown to stimulate different T lymphocyte populations
(CD4* vs CD8* respectively) in epithelial driven allogeneic MLR.
This stimulation can be blocked in both airway and intestinal stimu-
lators by preincubation of these cells with each of the epithelial-specific
mAbs L12, B9 or B39 (generated against lEC). In order to understand
the regulation of the molecules recognized by these mAb on AEC and
lEC, we utilized cell lines to aid in their characterization.

446

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

Method. Immunohistochemical staining was performed on forma-
lin fixed cells (Adherent A549 derived from alveolar epithelium; and
intestinal cell lines HT-29 and T-84). Immunoprecipitation of the epi-
thelial antigens recognized by these mAb was performed on '^'""I surface
labelled cell lysates lNP-40 37f/SDS 0.5%), and precipitated with either
anti-class I (W6/32), IgGl isotype control, or B9, L12, B39 mAbs, and
resolved on SDS-PAGE. Immunoblots were performed on unlabelled
cell lysates transferred to nitrocellulose after SDS-PAGE.

Results. Immunohistochemical staining of A549 revealed intense
surface staining for B39. L12 antibody staining was also found and
appeared to be upregulated after PMA exposure. Faint staining with
B9 mAb was detected, though no modulation was seen. A549 does not
express class II MHC by immunohistochemistry (mAb VG2.2), with or
without IFN-7 incubation, nor do A549 stimulate in allogeneic MLR.
HT-29 also express L12 and B39 epitopes, though no modulation was
seen with IFN-7, LPS, or PMA. '^^I Immunoprecipitation of A549 ly-
sates with L12 mAb revealed a 42-44 kD band. No immunoprecipi-
tates have been detected with B9 or B39 mAbs. In HT-29 immunopre-
cipitates, a L12 band at 45-55 kD was detected, as well as 69 kD B9
and 60 kD B39 bands. On immunoblot of A549 lysate, the LI 2 antibody
detected a 43 kd band, similar to the appearance of a immunoblot
prepared from intestinal cell lysate.

Conclusion. Airway and intestinal epithelial cell lines share a 43
Kd surface antigen (LI 2 Ag) which may be a novel restriction element
in cognate interaction between epithelial cells and lymphocytes. This
Eintigen differs from known accessory molecules (including MHC, LFA-
1, LFA-3, ICAM-1, 2, and B7/BB1) with regard to tissue distribution
and expression, functional characteristics, and immunoblot/immuno-
precipitation pattern. Expression of L12 antigen appear necessary but
not sufficient for stimulation of AEG driven MLR. Although L12 Ag
may be involved in cognate interaction with T lymphocytes, its pres-
ence on both AEG and lEG argue against the hypothesis that it is a
specific CDS ligand.

126. Elevated Intra-Abdominal Pressure and Intrinsic PEEP: Com-
mon Confounders of Occlusion Pressure Interpretation? Nelson JE,
Brown LK, Nierman DM, Kalb TH, Oropello JM. Divisions of Pulmo-
nary and Critical Care Medicine, and Surgical Intensive Care, Mount
Sinai Hospital, N.Y.

Interpretation of the pulmonary artery occlusion pressure (PAOP) is
complicated by changes in intrapleural pressure (Ppi). Two factors po-
tentially raising Pp,, elevated intra-abdominal pressure (Pab) and in-
trinsic PEEP (PEEP,), are of particular concern because they may go
unrecognized. The purpose of this study was to assess the prevalence of
these factors in patients with pulmonary artery catheters (PAG), and to
evaluate their effect on Ppi.

Methods. Twenty consecutive MICU patients with PAC were stud-
ied during assisted mechanical ventilation. P^b was measured using a
transurethral bladder catheter.^ PEEP, was measured (n = 19) by
respiratory inductance plethysmography.^ Ppi was measured (n = 9)
with a balloon esophageal catheter (CP-100 Monitor, BICORE). All
pressures were measured at end-expiration in semi-erect position.

Results. Thirteen of 20 patients had P^t ^ 10 mm Hg (range 2-19
mm Hg). Six had P^i, s 15 mm Hg. Four patients with P^b » 10 had no
clinical evidence of GI disease. Fourteen of 19 patients had detectable
PEEP,. Five of these had neither clinical evidence of flow limitation nor
Ve > 10 1/min. No patient had PEEP^ > 8. Neither PEEP,, P^b, nor
their sum was correlated with Ppi (p > 0.05).

Conclusions. In our sample, P^b * 10 occurred frequently, suggest-
ing that this is a common but under-recognized clinical entity. PEEP,
was also detectable in many patients, and not restricted to those with
obvious airflow limitation or high Vg, but occurred at modest levels.
Our inability to detect a consistent effect of these variables on Pp, may
reflect the role of such other variables as pulmonary or thoracic com-
pliance, or the relationship between bladder and subdiaphragmatic
pressures. Other studies have suggested that both Pj,b and PEEP, can
affect Ppi, and that Ppi affects both hemodynamics and PAOP interpre-
tation. Moreover, animal studies predict that PEEP, would cause even
greater increases in pulmonary capillary pressure than in PAOP.
Therefore, a clearer understanding of the interrelationship of these
variables in the ICU setting is warranted.

GGO # 91-529.

References

127. Characterization of the Potentiation of Bradykinin-Induced Con-
striction by an Inhibitor of Endopeptidase 24.15. E. N. Schachter, A.
Tsai, C. Gardozo, E. Zuskin, J. Godbold, N. Rienzi, M. Lesser. Mount
Sinai Medical Center and the VA Medical Center, New York, New
York and the Bronx, USA.

We have previously shown ( ARRD 143:A616, 1991) that the blocking of
a metallo-endopeptidase (EP24.15), that degrades bradykinin, by cFP-
AAY-pAB, enhances the bronchoconstrictor activity of bradykinin in
guinea pig trachea (GPT).

To evaluate the role of known modifiers of bronchoconstriction we
repeated our experiments under the following conditions: (1) GPT with
and without epithelium and (2) GPT in the presence of indomethacin 1
uM. We also studied the effect of the incubation time of GPT with
inhibitor, cFP-Ala-Ala-Phe-pAB (I 24.15). Four rings obtained from
each animal were initially stimulated with bradykinin 10 uM and
again, following treatment (3 minutes before the second challenge with
bradykinin) with 1) control solution 2) I 24.15 (10 uM) 3) Thiorphan (T)
(10 uM) 4) I 24.15 and Thiorphan. We also studied rings that had been
pretreated with indomethacin 1 uM and rings from which the epithe-
lium was removed. Finally we examined rings to which I 24.15 was
added 5, 15, 25 and 35 minutes before the second addition of bradyki-
nin.

Table 1 shows the percent increase in contractile response to
bradykinin (compared to baseline) after the addition of inhibitor.

Supported by NIOSH 5-ROl OH02493

With Without Indomethacin

epithelium epithelium pretreatment

+ 37%
+ 56%*
+ 110%t
+ 149%t

-14%
-1-25%
+ 65%
+ 31%

+ 97%*
+ 114%*
+ 424%t
+ 219%t

Iberti et al. Anesthesiology 1989; 70:47-50.
Hoffmann et al. Chest 1989; 96:613-616.

Control

124.15

T + I 24.15

*p< 0.05
tp < 0.01

The enhancing effect of I 24.15 can be seen in the presence of epithe-
lium but was not seen in its absence. Pretreatment with indomethacin
enhanced GPT reactivity to bradykinin in general. Finally enhance-
ment by I 24.15 was seen at 5' and 15' following incubation but not at

25' or 35'.

We conclude that EP24.15 in airway epithelium modulates brady-
kinin-induced contraction of GPT; this effect is non-specifically en-
hanced by suppressing prostaglandin sjmthesis.

128. Comparable Bronchodilator Effects of Albuterol and Ipratropium
in Mild Stable Asthma. Spiro P, Rienzi N, Schachter EN. Mount Sinai
School of Medicine, New York, NY.

Ipratropium bromide (IB) is an anticholinergic quaternary amine
which has known bronchodilator effects in patients with obstructive
lung disease. In asthma it is commonly held that ^-agonists are more
effective as bronchodilators than IB.

We developed a protocol to compare the relative bronchodilatory
effects of IB vs Albuterol in patients with stable asthma. We tested 9
subjects with a clinical history of asthma, average age 34.3, (range
21-57), without recent respiratory infection or the use of corticocos-
teroids in the last 4 weeks. All patients had an established bronchod-
ilator response (ATS criteria, 1991). Patients abstained from the use of
bronchodilator medications for at least 12 hours prior to testing. Sub-
jects received either IB (36 p.g) via two inhalations, or Albuterol (180
(xg) via two inhalations using standard meter dose inhalers. Spiromet-
ric measurements were obtained immediately before and ten minutes
after inhalation of the bronchodilator.

All patients responded to Albuterol but only 6 of 9 responded to IB.
Following Albuterol the following percent increases were noted, FVC:
11.5% ± 7 (mean ± SD), FEV1:23.2% ± 8. Following IB, among the
responders, the following percent increases were noted FVC: 10.2% ± 9
and for FEV1:24.2% ± 9. Comparison of these two responses by the
paired t-test showed no significant differences. We also correlated the
responses to Albuterol and IB with the age of the patient and the
baseline FEVl expressed as a percent of predicted and found no sig-
nificant correlation, (r = .173 and .234 Albuterol and r = .563 and .471
for IB)

We conclude that in a small group of stable young asthmatics with

Vol. 60 No. 5

ABSTRACTS

447

a positive response to Albuterol, the response to IB was frequent 66%
(6/9). When it occurred, the response to IB was as great as the response
to Albuterol at 10 minutes after the delivery of the medication.

Rheumatology

129. An Association between Refractory HELLP Syndrome and An-
tiphospholoid Antibodies during Pregnancy: a Report of Two Cases.
M. Omstein, J. Rand. Divisions of Rheumatology and Hematology, Mt.
Sinai Medical Center, New York, NY 10029.*

The HELLP Syndrome is a thrombotic microangiographic vasculopa-
thy that presents in pregnancy consisting of hemolysis, elevated liver
enzymes and low platelets. HELLP usually resolves with delivery of
the fetus without sequelae, and often is managed well with conserva-
tive care. There have been no prior reports of an association between
HELLP and the presence of elevated titers of anticardiolipin antibod-
ies.

We now report two cases of patients who each presented with
HELLP Syndrome that was refractory, despite delivery of the fetus,
corticosteroids, and anticoagulation. Both patients were found to have
elevated levels of anticardiolipin antibodies. Examination of skin and
placental pathology revealed diffuse deposition of fibrin with small
vessel thrombi, without evidence of vasculitis. Coincident with plas-
mapheresis, HELLP resolved in both patients.

We conclude that there may be an association between the pres-
ence of antiphospholipid antibodies and a protracted HELLP syndrome
in pregnancy; this may be a new manifestation of the antiphospholipid
syndrome. Plasmapheresis appears to benefit these patients.

130. Scleroderma in Association with Silicone Chin and Toe Implants:
a Report of Three Cases. Harry Spiera, Leslie Dubin Kerr. Division of
Rheumatology, Mount Sinai Medical Center, New York, NY 10029.

Scleroderma has recently been described in association with the use of
silicone-containing breast implants. We report our experience with
three scleroderma patients seen in association with the use of silicone
implants other than breast implants. Of these three patients, two had
the CREST variant of scleroderma and one patient had diffuse sclero-
derma (PSS). Two patients received silicone-containing chin implants.
The third patient received a silicone-containing hinge toe implant. The
patient with PSS died secondary to progressive cardiomyopathy despite
removal of the implant. The two CREST patients are currently clini-
cally stable with their implants in-situ.

These data extend the previous observations related to silicone
breast implants and scleroderma to other silicone-containing pros-
thetic devices. A history of exposure to any silicone-containing device
should be sought by physicians caring for scleroderma patients.

Clinical Abstracts

131. Hypothyroidism, Hypometabolism and Oxygen Consumption/De-
livery Dynamics. P. Spiro, D. Nierman. Division of Pulmonary and
Critical Care Medicine, Mount Sinai Medical Center, NY.

A case is presented in which invasive measurements of oxygen delivery
and consumption may have helped to distinguish true hypothyroidism
from the euthyroid sick state.

The patient was an 81-year-old female with HTN and mild OMS
transferred from another hospital for rehabilitation. Her previous hos-
pital course included a new diagnosis of Wegeners Granulomatosis.
Shortly after transfer she developed a RLL infiltrate on CXR. Despite
broad spectrum antibiotics, three days later she became hypotensive
and was intubated. Initial MICU evaluation revealed an obese, lethar-
gic woman, Temp^^^ of 33°C, with peripheral edema, sallow skin and
coarse hair. A PA catheter revealed a C.I. = 3.3 L/min, SVR = 1500
dynes ■ sec/cm'^, PCWP = 14 mm Hg, DO2 = 408 ml/min • m^, VO2 =
108 ml/min ■ m^, E.R. = .27 and a lactate of 1 mEq/L (Fig). Initial
thyroid function tests were consistent with sick euthyroid (T4 < 1
Hg/dl, TSH 3.2 (xU/ml). Unexpectantly, over the next eight days, de-
spite fluids, PRBCs, dopamine and dobutamine, her V02 as calculated

by the Pick method remained persistently low and fixed and multiple
lactates were normal over a wide range of DO^. On the 10th ICU day,
because of persistent hypometabolism, the patient was begun on 0.25
megs of synthroid. The normal initial TSH may have been secondary to
the exogenous administration of dopamine. The V02 and ER(02) pro-
gressively improved and she was stable off pressors. TFTs 3 days after
starting synthroid revealed a paradoxical rise in TSH. After eight days
on sjTithroid she developed hemodynamics of septic shock and, despite
aggressive treatment, died 9 days later.

Oxygen Delivery/Consumption Data

0«yg«n Consumption (ml/min/m2) Extraction Ratio (%)

ol — ' — ' — ' — I — ' — ' — ' — I — ' — ' — ' — I — ' — ' — ' — I — — ' — I — '— ' — '0
12/14 12/18 12/22 12/2« 12/30 1/3

Data

Post-mortem exam showed a markedly atrophic thyroid gland.
Persistent and unexplained hypometabolism may help to differentiate
hypothyroidism from ESS.

132. Thyrotropin-Secreting Pituitary Macroadenoma: Endocrine, Bio-
chemical, Morphological Studies, and Therapeutic Approach. M. Al-

varez-Marfany, S. Morgello, K. Post, A. C. Levine. Divisions of Endo-
crinology, Neurosurgery, and Pathology. Mount Sinai Hospital, NY.*

A 40-year-old male with a chief complaint of visual difficulty was re-
ferred for evaluation of a pituitary mass. The clinical evaluation was
remarkable for visual loss in the right eye, headaches, profuse sweat-
ing, heat sensitivity, insomnia, palpitations, and weight loss. Right
temporal hemianopsia and a small diffuse goiter were noticed on phys-
ical exam. MRI scan showed a very large sella suprasellar mass lesion
with eccentric extension towards the right. The ophthalmologic eval-
uation was consistent with right optic nerve compression. High serum
levels of T4 (23.9 pig/dL) and T3 (498 ng/dL) associated with a supran-
ormal serum TSH (7.8 |xU/mL) were detected. Upper normal serum LH
levels (21 mlU/mL) were also noted with normal levels on the remain-
der of the anterior pituitary hormones.

These findings indicated the presence of hyperthyroidism due to
inappropriate secretion of TSH, whose neoplastic origin was clinically
evident and documented by imaging studies. The diagnosis was con-
firmed by elevated a-subunit level (21.1 ng/mL), and a-subunit/TSH
molar ratio (24.2). Subsequently a transsphenoidal adenectomy with
subtotal removal of the adenoma was performed, followed by radiother-
apy for residual tumor. Immunohistochemical staining of the adeno-
matous tissue removed at surgery was positive for TSH and LH. Fol-
lowing the surgery, TSH levels fell transient, the hyperthyroidism
persisted. The patient is presently being treated with propylthiouracil.
Somatostatin analog therapy is also being considered.

TSH-secreting pituitary tumors are exceedingly rare. They are
generally large and invasive. The prognosis is inversely related to the
size of the tumor at the time of diagnosis. Hyperthyroxinemia with
inappropriately elevated or normal TSH, in the presence of a pituitary
tumor and an elevated a-subunit are the diagnostic hallmarks. Con-
comitant secretion of LH is rare. Although surgical resection is the first
line of therapy, many of these tumors are large and incomplete resec-
tions are common. Radiotherapy is administered when the resection
has been incomplete, but the efficacy of radiotherapy has not been well
documented. Therefore adjunctive medical therapy is essential. Clini-
cal improvement and significant tumor reduction have been obtained
with somatostatin analog therapy.

133. Severe Status Asthmaticus Requiring Inhalational Anesthesia.
M. Forman, MD, T. Kalb, MD. Division of Pulmonary and Critical Care
Medicine, Mount Sinai Medical Center, NY.

A 49-year-old male was admitted to the Mount Sinai Medical Center
Medical Intensive Care Unit with severe bronchospasm. Conventional

448

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

treatments including intravenous steroids, beta-agonists, and mechan-
ical ventilation were ineffective in this case. Inhalational anesthesia
provided ventilatory support, and an airflow meter was used to mea-
sure exhaled tidal volumes as a gauge to therapy. Bronchodilation was
achieved and the patient recovered.

134. Clinical Case Presentation: Jaundice in a Patient with Prostate
Cancer. C. Frissora-Rodeo, A. Rosman, A. Marshall, B. Reiter, and F.
Paronetto. The Mount Sinai Medical Center, Department of Medicine.

A 72-year-old man was admitted because of jaundice and pruritus.
Three days prior to admission the patient noted pruritus, jaundice, and
tea-colored urine. He denied fever, abdominal pain, nausea, and vom-
iting. He was diagnosed with adenocarcinoma of the prostate seven
years prior to admission. He was treated with radical prostatectomy at
that time and serial imaging studies revealed no evidence of metastatic
disease. He was well until three months prior to admission when he
underwent an above-the-knee amputation after a motor vehicle acci-
dent. He required transfusion of packed red blood cells postoperatively.
A repeat bone scan revealed metastatic disease in the spine and the
iliac crest. Abdominal CT scan revealed enlarged retroperitoneal
lymph nodes and the liver appeared normal. He was treated with or-
chiectomy and flutamide. The medical history was significant for dia-
betes mellitus, hypertension, and an inferior wall myocardial infarc-
tion. He formerly was a bartender and consumed 16 ounces of whiskey
per day. His last drink was thirteen years ago. Medications included
flutamide, hydrochlorthiazide, NPH insulin, and salicylic acid.

On examination the patient was awake and alert. The tempera-
ture was 38 degrees, the pulse was 76, and the respirations 14. The
blood pressure was 120/70. The skin was jaundiced and the sclerae
icteric. The lungs were clear, and the heart sounds were normal. The
abdomen was mildly obese, and there were no caput medusae or angi-
omas. The bowel sounds were normal and the abdomen was soft with
mild right upper quadrant tenderness. There was no organomegaly,
mass, or evidence of ascites. Rectal exam revealed no masses and light
stool which was negative for occult blood. Laboratory studies revealed:
white cell count 7,200/mm'' with a normal differential; hemoglobin, 10
gmydL; platelets 228,000; total bilirubin, 13 mg/dL; direct bilirubin 8
mg/dL, aspartate aminotransferase 305 U/L; alanine aminotransferase
443 U/L; alkaline phosphatase 3054 U/L; gamma-gutamyl transpepti-
dase, 669 U/L; and prothrombin time, 11 seconds. Hepatitis serologies
were negative. Urinalysis was significant for bilirubinuria. Five weeks
prior the bilirubin, ALT, and AST were normal. Abdominal sonogram
and abdominal CT scan revealed no evidence of liver metastasis, bile
duct dilatation, or gallstones. Liver biopsy revealed significant intra-
hepatic cholestasis, feathery hepatocytes containing granules of bile
pigment, and focal necrosis of hepatocytes surrounded by a prominent
accumulation of neutrophils and mononuclear cells. These features
were interpreted as being drug induced. Flutamide was discontinued
on admission. The total bilirubin decreased to 2 mg/dL one month after
the flutamide was discontinued.

The patient remains well, and is living at home.

135. Clinical Case Presentation: Sudden Death in a 60-Year-Old Ten-
nis Player. C. Frissora-Rodeo, B. Fyfe, S. Rodeo. The Mount Sinai
Medical Center, Department of Medicine and Department of Pathol-
ogy-

A 60-year-old man consulted an orthopedic surgeon with a complaint of
right neck pain. He was in good health until ten days prior when he
developed a nonproductive cough and myalgia. Two days prior, the
patient played nine sets of tennis and was struck by a tennis ball on the
right mastoid process. For 24 hours following this injury he complained
of increasing sharp right neck pain, with occasional radiation to the
right scapula and hand. The medical history was significant for occa-
sional viral syndromes and cervical spondylosis. The surgical history
included complete resection of well-differentiated papillary carcinoma
of the right thyroid and intussusception in infancy. Otherwise he was
in extremely good health, and engaged in competitive tennis regularly.
The patient was an entrepeneur chemical engineer, and served in the
Korean War. There was no history of obesity, hypertension, tobacco
use, hypercholesterolemia, or diabetes. He consumed three glasses of
wine approximately four times per week. Three brothers developed
coronary artery disease at approximately age 55. Medications were:
levothyroxine .175 mg daily and ibuprofen 400 mg p.o. every eight
hours as needed.

Upon examination he was in moderate discomfort. Physical exam-
ination was significant for cervical muscle spasm. Plain X-ray revealed
degenerative changes in the cervical spine. He was treated with a

cervical collar, naproxyn, acetaminophen, tylenol, and cyclobenza-
prine. An MRI was scheduled. The patient continued to complain of
neck pain which was partially relieved by the above medications. He
noted myalgia, a nonproductive cough, pharyngitis, arthralgia, an-
orexia, one episode of nonbloody vomiting, and occasional tachypnea
when the neck pain was severe. During the MRI he was tachypneic but
otherwise had no orthopnea, palpitations, or edema.

Four days after the office visit, during a telephone call to his or-
thopedic surgeon, the patient complained of malaise, increasing right
arm and neck pain, and decreased ability to move his right arm. The
surgeon noted mild confusion which he attributed to cyclobenzaprine,
which was replaced with diazepam. The patient returned home, did not
ingest diazepam, and fell asleep. Several hours later he had labored
breathing and a cardiopulmonary arrest. EMS found the patient to be
in asystole and then in ventricular fibrillation. ACLS protocol was
initiated. The patient was dead on arrival to the emergency room.

Post-mortem examination revealed: white blood cell count
12,500/mm^ with neutrophils 66%, bands 7%, lymphocytes 19%, mono-
cytes 8%; normal thjToid function tests; and Coxsackie Bl, B3, and B6
titres of 1:32. Polymerase chain reaction for Cocksackie B viral parti-
cles was negative. The heart weighed 450 g, the left ventricular wall
thickness was 1.5 cm, and the valves were normal. Microscopic exam
revealed focal myocarditis with infiltration of histiocytes and occa-
sional eosinophils, predominantly in intersitial tissue planes. These
findings were interpreted by two experts as a hypersensitivity reaction.
The pathology was not consistent with a viral myocarditis, due to lack
of lymphocytic infiltration.

136. Clinical Case Presentation: Syncope after Cocaine Use. Christine
Frissora-Rodeo, MD and Alain Zilkha, BA. Department of Medicine,
The Mount Sinai Medical Center, New York, NY.'''

A 61 -year-old man was admitted for the evaluation of shortness of
breath and palpitations. He was well until two weeks prior to admis-
sion when he had a syncopal episode while dancing. He complained of
palpitations prior to the event. No seizure activity was noted. He had
consumed alcohol and used nasal cocaine approximately three hours
prior to the syncopal episode. He subsequently developed progressive
dyspnea, low grade fever, and was treated at an outside clinic with
Erythromycin. He then noted orthopnea, palpitations, and paroxysmal
nocturnal dyspnea. There was a history of rheumatic fever at age 18.
He reported thirty sexual partners in the last three years and used
cocaine biweekly. He was never tested for antibody to HIV. He was
employed in a textile factory.

On physical examination the patient appeared in severe respira-
tory distress. The temperature was 37.6 degrees centrigrade, the pulse
was 136 and irregular, and respirations were 46. The blood pressure
was 130/70. The inspiratory drop in systolic blood pressure was three.
Venous pulsations were visible at the level of the mandible. There were
decreased breath sounds and bilateral rales at the lung bases. Diffuse
wheezing was also present. The heart rate was tachycardic and irreg-
ular. A II/VI diastolic murmer was heard at the apex in the left lateral
decubitus position, and a III/VI holosystolic murmur was heard at the
apex which radiated to the axilla. The abdomen was soft, and there was
mild edema in the extremities. Laboratory data revealed: white blood
cell count, 8,700/mm^ with a normal differential; hemoglobin, 13.3
gm/dL; platelets 288,000, potassium 4.6 meq/liter; creatinine 1.1
mg/dL; creatinine kinase 89 U/L; lactate dehydrogenase 412 U/L; and
normal thyroid function tests. EKG revealed atrial fibrillation at 130
and right bundle branch block pattern. HIV antibody was negative.
Echocardiogram revealed moderate mitral stenosis, mild mitral regur-
gitation, a markedly dilated left atrium, and a right ventricular sys-
tolic pressure of 70 mm. The right ventricle was dilated with decreased
function. Cardiac catheterization after diuresis revealed a mitral valve
area of 1.4 cm^, moderate mitral regurgitation, and a transmitral gra-
dient of 8 mm Hg. The pulmonary capillary wedge pressure was 30 mm
Hg before and 50 Hg mm after dye injection. The left ventricular func-
tion and coronary arteries were normal.

The patient was treated with furosemide, atenolol, heparin and
Coumadin. He has returned to work and is well.

137. Maffucci's Syndrome with Portal Colopathy: Case Report. James
George, M.D. Division of Gastroenterology, Mount Sinai Medical Cen-
ter, NY, NY.

The patient is a 45-year-old white female with a diagnosis of Maffucci's
syndrome, characterized by an abnormal growth pattern with elonga-
tion of the left side of her body, multiple bony deformities and enchon-
dromatosis. In May 1992 the patient went to an outside hospital with

Vol. 60 No. 5

ABSTRACTS

449

an acute massive lower gastrointestinal bleed and underwent emer-
gent subtotal colectomy with ileorectal anastomosis. The patient was
told she had ulcerative colitis and was discharged. The patient came to
Mount Sinai in March 1993 with bright red blood per rectum. Upper
endoscopy revealed diffuse gastric varicies. Colonoscopy revealed an
inflamed rectum with granularity and intense friability of the rectal
mucosa, a normal ileum and external hemorroids. A duplex sonogram
revealed cavernous transformation of the portal vein, with large var-
icies in the hilum and stomach. Angiography revealed cavernous trans-
formation of the portal vein with dilated veins surrounding the rectum.
The patient underwent a gastroepiploic to internal iliac vein potosys-
temic shunt for intractable bleeding and her portal venous pressure
dropped from 30 mm to 15 mm of mercury. Preoperatively the patient
lost 25 units of blood and postoperatively 2 units of blood.

This case represents an unusual manifestation of Maffucci's syn-
drome with portal colopathy, a condition that can be misdiagnosed as
ulcerative colitis.

138. Fever, Night Sweats, and Weight Loss in a 55-Year-Old Man.
Jonas Leibowitz, MD, Eric Grubman, MD, William Suozzi, MD.

A 55-year-old black male was admitted to the hospital complaining of
several weeks of anorexia, exertional dyspnea, low grade fevers, night
sweats, fatigue, and significant weight loss. The patient had been well
until several weeks before admission, when he noticed anorexia with-
out associated nausea or vomiting. He complained of three weeks of
night sweats and arthralgias. He also noted a 35-lb weight loss of the
previous several months. On the day of admission, the patient con-
sulted his physician with the above complaints and described signifi-
cant exertional dyspnea. The patient worked as an accountant. His
wife and child were well. There was a 30-pack year history of cigarette
smoking and the patient ingested alcohol until ten years earlier when
he was hospitalized for alcoholic pancreatitis and required blood prod-
uct transfusions. Many years earlier he was hospitalized for bacterial
pneumonia, and six years ago he suffered a CVA with mild residual
weakness in his left upper extremity. The patient's father died from a
pulmonary embolus at the age of 38 and his mother was alive and had
previously had a stroke. The patient had a positive tuberculin reaction
and was treated with INH many years ago. His medications consisted
of a thiazide diuretic and Coumadin. The remainder of the history was
unremarkable.

On physical examination, the patient was febrile to 102 degrees.
The pulse was 104 and respirations were 18. The blood pressure was
120/74. The patient was thin and appeared younger than his stated
age. No cutaneous lesions or lymphadenopathy was found. The head,
neck, heart, and abdomen were normal. The lung exam revealed dif-
fiase end-expiratory wheezes. The rectal, genital, extremity and periph-
eral pulse exam were normal. The neurological exam revealed mild left
upper extremity weakness. The hematocrit was 30.6 percent; the
white-cell count was 3900, with 80 percent neutrophils and 12 percent
lymphocytes; the platelet count was 180,000, and the ESR v/as 85 mm
per hour. The serum chemistries were normal. The arterial blood gas
revealed a pH of 7.46, a pC02 of 34 mm Hg and a p02 of 84 mm Hg.
Radiographs of the chest revealed no lymphadenopathy and changes
consistent with chronic lung disease. A ventilation-perfusion scan was
indeterminate and possibly consistent with severe lung disease. A pul-
monary angiogram revealed no evidence of a pulmonary embolus. The
patient was anergic.

The patient was started on an inhalational bronchodilator. A CT
scan of the chest and abdomen with intravenous contrast revealed
paratracheal and subcarinal lymphadenopathy, bullous changes at the
bases, a small nodular density in the right lower lobe, and a small
pericardial effusion. Multiple blood cultures were sent, all of which
remained negative. The patient remained febrile. Iron studies were
consistent with anemia of chronic disease with a markedly elevated
ferritin. A test for human immunodeficiency virus (HIV) antibody was
negative and lymphocjrte studies revealed a normal CD4:CD8 ratio. An
ANA and anti-ds DNA was positive. Additionally, anticardiolipin an-
tibodies and a rheumatoid factor were positive and complement levels
were low.

A diagnosis of SLE was made. The patient was started on tapering
doses of prednisone, as well as plaquenil. His symptoms gradually re-
solved. The patient was discharged with a diagnosis of SLE and chronic
lung disease.

139. A Case of Acute Pulmonary Renal Syndrome. Jill Ritter and
Leslie Dubin Kerr. Division of Rheumatology, Mount Sinai Medical
Center, New York, NY.

A 42-year-old Hispanic woman consulted a physician for renal failure
and shortness of breath with a vague prolonged history of constitu-
tional symptoms, anemia, and synovitis. She denied fever, cough, chest
pain, or hemoptysis. Her physical exam on admission: VS: BP =
120/70, HR = 110, RR = 34, afeb. Lung: exam revealed bilat rales and
rhonchi throughout; Cor: tachycardia without murmurs. Extremity
exam showed palpable purpura on one fingertip, no edema or synovitis.
Neuro exam was nonfocal. Lab data of note: WBC = 4.7, Hb = 5.1, Hct
= 14.8. Na = 141, K = 3.9, CI = 115, HCO3 = 16, BUN/cr = 60/7.2.
Urine sediment showed RBC casts. There was no evidence of hemoly-
sis. CXR showed diffuse bilat infiltrates, and ABG on RA = 7.35/28/68.
Serologies included: ANA 1:640, anti-DS-DNA neg., anti-Smith anti-
body neg., anti-glomerular basement membrane antibody neg., C3 low
at 67, C4 normal. P-ANCA positive 1:320; C-ANCA negative.

Patient was admitted to MICU with acute pulmonary renal syn-
drome; the differential diagnoses include Goodpasture's syndrome, We-
gener's granulomatosis, or systemic lupus. Pt. received 2u PRBC; Me-
drol Igm IVSS QD x 3d, and plasmapheresis pending diagnostic
workup. Renal biopsy showed crescentic necrotizing glomerulonephri-
tis, and oral cyclophosphamide was started. Plasmapheresis was dis-
continued when immunofluorescence of renal bx. showed only small
complement deposits, no linear deposits c/w Goodpasture's syndrome.
Electron microscopy confirmed pauci immune necrotizing glomerulo-
nephritis c/w Wegener's granulomatosis.

Patient was discharged on the eleventh hospital day on prednisone
60mg QD and Cytoxan lOOmg PO QD. Her creatinine decreased to 3.5,
and CXR showed almost complete resolution of bilateral infiltrates.

140. A Case of Connective Tissue Disease Complicated by Vasculitis,
Gangrene, and Tamponade. Jill Ritter, Leslie Dubin Kerr. Division of
Rheumatology, Mt. Sinai Medical Center, New York, NY 10029.

A 20-year-old black woman had a 4-year history of undefined CTD
manifested by cutaneous scleroderma, active myositis, Raynaud's phe-
nomenon, necrotic hands and feet, blurred vision, and SOB. Admission
physical exam was significant for cachexia VS: BP = 120/70 w/o pul-
sus; HR = 130; RR = 24, afeb. Wt = 65 lbs., ht = 5'6". HEENT:
alopecia, oral ulcers. Fundoscopic exam revealed inflammatory infil-
trates with pale optic nerve on left. Cor: tachycardia with a gallop, pos.
JVD. Lung: bilat. rales halfway up. Ext. exam revealed black necrotic
dry gangrene 18/20 digits to PIP joints, left wrist drop, and diffuse
sclerodermatous skin with pressure ulcers. Lab data of note included
Hct = 22%, elev. LDH, and Coombs pos.; normal creat.; U/A: moderate
blood, no protein; CPK = 79; albumin = 1.9. Numerous autoantibodies
were positive including: ANA, anti-DS-DNA, anti-RO, anti-Scl-70,
Anti-RNP antibody, anticardiolipin antibodies of IgG and IgM classes.
C3 level was decreased at 65. CXR showed water-bottle-shaped heart
and EKG revealed sinus tachycardia with low voltage. Echocardio-
gram confirmed pericardial tamponade, and GBPS showed EF = 20%.

The patient was treated with high dose IV steroids without im-
provement. She required a pericardial window for refractory pericardi-
tis, and Cytoxan was started for vasculitis. She also received IV gamma
globulin and plasmapheresis. Our clinical impression was mixed con-
nective tissue disease complicated by vasculitis of the retina, digital
arteries and peripheral nerves, cardiomyopathy, tamponade, throm-
bosis secondary to antiphospholipid syndrome, and malnutrition. After
a stormy four-month course complicated by persistent exudative drain-
age from pleural/pericardial drains, catheter sepsis, esophagitis, GI
bleeds, she is currently markedly improved. There is no reaccumula-
tion of pleural/pericardial effusions. Repeat GBPS shows EF = 36%, no
deterioration of vision, and resolution of wrist drop. Digital gangrene
has not progressed; necrotic tissue is autoamputating. Her weight is
now 88 lbs., and albumin, 3.8.

141. Ischemic Ulcerations of the Rectum in a Patient with Primary
Antiphospholipid Syndrome. Jill Ritter M.D. and Tony Weiss M.D.
Divisions of Rheumatology and Gastroenterology.

A 53-year-old black woman had a history of end stage renal disease on
hemodialysis for one year. One day after her repair of a clotted graft
she developed monocular blindness of the right eye. Ophthalmalogic
exam was suggestive of ischemic retinopathy. The patient had an ESR
of 100 and was started on high dose steroids for presumed temporal
arteritis. Transthoracic echocardiogram and carotid Dopplers were
negative. CT scan of the head revealed a right occipital infarct. A
temporal artery biopsy was negative and steroids were tapered. The
patient was admitted with fluctuation in her mental status while on 40
mg qd of prednisone. She was febrile to 100.7 on admission but physical
exam was otherwise unremarkable. Repeat CT scan revealed a possible

450

THE MOUNT SINAI JOURNAL OF MEDICINE

October 1993

left caudate lobe infarct. EEG and lumbar puncture were unremark-
able. Cultures of blood and cerebrospinal fluid were negative. Steroid
related delirium was suspected and the taper was continued. On the
tenth hospital day she passed a large amount of bright red blood per
rectum. She became hypotensive, requiring fluid resuscitation, trans-
fer to intensive care, and transfusion of five units of PRBCs. An upper
endoscopy revealed superficial gastric erosions. Colonoscopy was re-
markable for ulcerations of the distal rectum and necrotic-appearing
tissue. Biopsy showed focal superficial necrosis suggestive of ischemic
mucosal injury.

Serologic studies revealed an IgG antiphospholipid antibody titre

of 6.8 (nl < 3) and a negative ANCA. The angiographic studies of the
mesenteric and iliac vessels were normal. The patient was treated ini-
tially with a rectal suspension of carafate and subsequently with in-
travenous heparin. On day 19 a repeat endoscopic exam revealed sig-
nificant healing of the ulcers. She remained without evidence of
recurrent bleeding and continues on long-term anticoagulation.

Gastrointestinal complications of primary antiphospholipid syn-
drome are rare and have previously been reported to result in intesti-
nal ischemia and infarction. We believe that this case is important in
showing that ulcerations of the rectum can be seen with this disease
and may present with significant bleeding.

The Department of Rehabilitation Medicine
The Mount Sinai School of Medicine (CUNY)

The Page and William Black
Post-Graduate School of Medicine

The Rehabilitation Medicine Services
Veterans Affairs Medical Center, The Bronx, NY

sponsor the first international course on

myofascial
PAIN

diagnosis & treatment

emphasizing practical clinical management

course: December 8-11, 1993
workshop: December 12, 1993
The Mount Sinai Medical Center
100th St. & Madison Ave., New York, NY
Stern Auditorium

course director: Andrew A. Fischer, MD, PhD, Chief, Rehabilitation Medicine Service, VAMC, Bronx, NY
faculty: Janet Travell, MD, Hans Kraus, MD, and other leading authorities

Fees:

course: $390 physicians

$290 others
workshop: $120 physicians
$100 others

CME credits available

designed for pain specialists, including physiatrists,
orthopedists, anesthesiologists, rheumatologists,
neurologists, physical and occupational therapists,
chiropractors, and myotherapists

discounted hotel and travel arrangements for registrants: Zenith Travel (212) 241-9447

for information call The Page and William Black Post-Graduate School of Medicine, Mount Sinai
School of Medicine, Box 1193, One Gustavo L. Levy Place, New York, NY 10029

(212) 241-6737

NOVEMBER 1993
Annual Orthopaedic In-Tralning Exam Review

Michael M Lewis, MD

Symposium on Venous Diseases

Harry R Schanzer. MD
Robert A Nabatoff, MD

Chemotherapy Foundation Symposium XI
Innovative Cancer Chemotherapy for Tomorrow

Holiday Inn Crowne Plaza
Ezra M. Greenspan, MD

Evening Seminar: What's New for Cancer
Patients?

Holiday Inn Crowne Plaza
Ezra M. Greenspan, MD
Mary-Ellen Siegel. MSW

Eastern Group Psychotherapy Society Annual
Conference

Robert Friedman, PhD
Hillel Swiller, MD

4 Soc for Clinical and Experimental Hypnosis:
Introductory Clinical Hypnosis Workshop

Marianne S Andersen, PhD
Stephen Snyder, MD

The Dean's Lecture Series, 4 PM Hatch Aud
Manipulating the Mouse Embryo: From Cells
to Genes

Janet Rossant, PhD, Mt Sinai Hosp of Toronto

DECEMBER 1993

Training In Child Abuse Recognition &
Reporting

6 PM Hatch Aud

Katherine Teets Grimm, MD

repeated 2/1, 3/29, 6/28

The Dean's Lecture Series, 4 PM Hatch Aud
Using an Ancient Enzyme to Regulate the
Fate of RNA: How Eukaryotes Sense and
Control a Toxic Nutrient

Richard D. Klausner, MD, Natl Institutes of Health

Myofascial Pain: Diagnosis & Treatment

Andrew A. Fischer, MD

JANUARY 1994

Symposium on Women's Health Care Issues
1 1- In the 90's: For Primary Care Providers

I Michael Brodman, MD
Martin E. Goldman, MD
Stacey E. Rosen, MD

The Dean's Lectures Series, 4 PM Hatch Aud
Memory and the Mind

Patricia Goldman-Rakic, PhD
Yale University School of Medicine

Brachytherapy for Prostate Cancer

Nelson N Stone, MD

Queens Hospital Center Internal Medicine
Board Review Course

Wednesday evenings through June 1994
Laguardia Marriott Hotel
Fred Rosner, MD

American Academy of Psychiatry and the Law
Conference

Karl M Easton, MD

2> International Trauma Anesthesia & Critical
Care Soc: 1994 Trauma Anesthesia Update

Lake Tahoe, California
Kenneth J, Abrams, MD

28 40th Comprehensive Sinus Surgery Course

William Lawson, MD, DDS
Hugh F Biller, MD

30 Soc for Clinical & Experimental Hypnosis:
Advanced Clinical Hypnosis Workshop

Marianne S Andersen, PhD
Stephen Snyder, MD

MOUNT SINAI SCHOOL OF MEDICINE

The Page and William Black
Post-Graduate School of Medicine
Of Mount Sinai School of Medicine
Of The City University of New York

Continuing
l\1edical

Education
Courses

autumn 1993-spring 1994

FEBRUARY 1994

2-6 Geriatrics Board Review Course

Robert N Butler, MD, Judith Ahronheim. MD,
Gabriel Gold, MD

5-6 Advanced Hypnosis in Behavioral Medicine.

Psychosomatic Disorders & Consultation Liaison
Psychiatry

Harold J Wain, PhD, Steven Snyder, MD

9 The Dean's Lecture Series, 4 PM Hatch Aud

Molecular Properties of Voltage Gated Ion
Channels

William A Catterall, PhD

Univ of Washington School of Medicine

MARCH 1994

4 Issues In Medical Ethics 1994: Scientific Medicine,

Alternative Care, and Third-Party Payers

Rosamond Rhodes, PhD

9 The Dean's Lecture Series, 4 PM Hatch Aud

Mechanism & Regulation of Protein Transport
Early In the Secretory Pathway

Randy Schekman, PhD

University of California at Berkeley

5-12 Skull Base Surgery

Chandranath Sen, MD & Peter Catalano, MD

Head & Neck Reconstruction

Mark Urken, MD
Hugh F. Biller, MD

18 25th Annual Communications Disorders

Conference

Arnold Shapiro

7-8

for information, please call (212) 241-6737

The Page and William Black
Post-Graduate School of the
Mount Sinai School of t^edicine (CUNY)
is accredited by the Accreditation Council
for Continuing f^edical Education (ACCME)
to sponsor continuing medical education
for physicians

unless otherwise indicated, all courses
held on Mount Sinai campus at
100th Street & Madison Avenue, New York City

20-22

22-23

APRIL 1994

Alternate Treatments for Localized Carcinoma
of the Prostate: Radical Perineal Prostatectomy,
Brachytherapy, Cryosurgery

Nelson N Stone, MD

The Dean's Lecture Series, 4 PM Hatch Aud
New Concepts in Steroid Receptor Activation

Bert O Malley, MD

Baylor College of Medicine

Aortic Surgery Symposium IV

New York Hilton & Towers
Randall B Griepp, MD

Thyroid Disease and the Eye

Arthur L. Millman, MD

MAY 1994

7 Conference on Mediastinal Diseases

Steven M Keller, MD
Paul A Kirschner. MD

8 Echocardiography Workshop

Martin E. Goldman, MD

9-13 Consultant's Course In Cardiology

Louis E Teichholz, MD
Richard Gorlin, MD
Simon Dack, MD

11 The Dean's Lecture Series, 4 PM Hatch Aud

Signal Transduction by Receptors that
Regulate Tyrosine Phosphorylation

Joseph Schlessinger, PhD
NYU School of Medicine

JUNE 1994

8 The Dean's Lecture Series, 4 PM Hatch Aud

Receptor Protein-tyrosine Kinases and
Phosphatases and Their Substrates

Tony Hunter, PhD

The Salk Institute For Biological Studies

Required Forms for Authors:

AUTHORSHIP RESPONSIBILITY, FINANCIAL DISCLOSURE, AND ASSIGNMENT OF COPYRIGHT

Each author must read and sign (1) the statement on authorship responsibility; (2) the statement on financial disclosure; and (3) either the statement
on copyright transfer or the statement on federal employment. If necessary, photocopy this document to distribute to coauthors for their signatures.
Please return all copies to the address below.

I . .\uthursliip Responsibility

1 certify that I have participated sufficiently in the conception and
design of this work and the analysis of the data (where applicable), as
well as the writing of the manuscript, to lake public responsibility for it.
I believe the manuscript represents valid work. I have reviewed the final
version of the manuscript and approve it for publication.

Author(s) Si^ature(s)

Neither this manuscript nor one with substantially similar content under
my (our) authorship has been published or is being considered for
publication elsewhere, except as described in an attachment.

Furthermore, I attest that I shall produce the data on which the manu-
script is based for examination by the editors or their assignees should
they request it.

Date Signed

2. Financial Disclosure

Author(9) Signature(s)

sultancies, stock ownership, honoraria), except as disclosed in an at-
tachment.

Any financial project support of this research is identified in a note on
the title page of the manuscript.

Date Signed

3. Copyright

AulhoHs) Signature(s)

thor(s), thereby transferring, assigning, or otherwise conveying all copy-
right ownership, including any and all rights incidental thereto, exclu-
sively to The Mount Sinai Journal of Medicine.

Date Signed

LS Federal Employees: 1 was an employee of the LS federal govern-
ment when this work was conducted and prepared for publication; there-

Author(8) Signature(s)

fore, it is not protected by the Copyright Act and there is no copyright;
thus, ownership cannot be transferred.

Date Signed

Return the original signed form to Sherman Kupfer, M.D., Editor, The Mount Sinai Journal of Medicine, Box 1094, 50 E. 98th Street,
New York, NY 10029. Retain one copy for your files. You may make as many photocopies as you need if your paper has more than
three authors.

Instructions for Authors

THE MOUNT SINAI JOURNAL OF MEDICINE, established in 1934 as The Journal of the Mount Smai Hospital, is a peer-re-
viewed general medical journal. It is indexed or abstracted in Index Medicus, Current Contents, Science Citation Index, Hospital
Literature Index, International Nursing Index, Excerpta Medica, Chemical Abstracts, Biological Abstracts, and major databases.
Theoretical and basic science articles for the clinician, reviews, reports of clinical or research experience, articles on any subject
affecting the practice of medicine, short notes in any specialty, book reviews, letters, and articles by medical students are
welcome. The Journal is a participant in the Agreement on Uniform Requirements for Manuscripts Submitted to Biomedical
Journals. The Journal aims at responding to authors of manuscripts within 8 weeks of date of submission.

Authorship Responsibility, Financial Disclosure, and Assign-
ment of Copyright. Each author is required to sign the three
statements on the accompanying page.

Manuscript Preparation. Submit manuscripts in triplicate,
including three sets of illustrations. Manuscripts must be
typewritten, double-spaced throughout (including refer-
ences, legends, and tables), on one side of 8^/2 x 11-in. white
bond paper, with generous margins. Number pages consecu-
tively. Begin a new page for title page, abstract, first page of
text, references, each table, and legends. Computer-generated
typescripts must be double-spaced and must be easy for ed-
itors and typesetters to read. Computer tapes or disks: consult
managing editor.

Key Words. Supply 3 to 10 key words or phrases at the bottom
of the title page, for use in indexing the article. Use Med-
ical Subject Headings from Index Medicus.

Units of Measurement. Use SI units, giving range or interval
of normal values.

Illustrations. Photographs, line drawings, graphs, and charts
should increase understanding of the text. Line drawings,
graphs, and charts should be professionally prepared. All il-

1. Nadelman RB, Wormser GF. A clinical approach to Lyme
disease. Mt Sinai J Med 1990;57:144-156.

2. International Committee of Medical Journal Editors. Llni-
form requirements for manuscripts submitted to biomedical
journals. N Engl J Med 1991;324:424-428.

When a manuscript is accepted for publication, you will be
asked to confirm the accuracy of all references.

Reprint Orders. A reprint order form is sent to the corre-
sponding author with proofs; return it to the editorial ofTice.

THE

MOGNT SINAI JC
OF MEDICINE

tliri^llillll^irjWl.flM'ft'-EDICAL

3 4805 0364599 7

.RENTER

Forthcoming:
Grand Rounds

Gene Therapy for Immunodeficiency and Cancer

R. Michael Blaese
National Institutes of Health

Neonatal Herpes Simplex Virus Infections:
Natural History, Pathogenesis, and Preventability

Richard J. Whitley

University of Alabama School of Medicine

Cytokine Gene Therapy

Bernd Gansbacher

Memorial Sloan-Kettering Cancer Center

The Function of the p53 Tumor Suppressor
Gene and Its Clinical Correlates
Arnold Jay Levine
Princeton University

Mechanisms of Autoantibody Production and
Their Role in Disease

Keith Elkon

The Hospital for Special Surgery

Pathway for Water Absorption in
Isosmotic Transporting Epithelia

Guillermo Whittembury

Instituto Venezolano de Investigaciones Cientificas

Theme Issues

NOVEMBER 1993

Geriatrics Education, Clinical
Services, and Research:
A Decade of Experience

Editors: Myron Miller, Rein Tideiksaar,
and Carol Capello

JANUARY 1994

Cervical Disk Disease:
Controversies in Neurosurgeiy

Editors: Kalmon Post and Martin H. Savitz

MARCH 1994

Update on Sleep Disorders

Editor: Lee K. Brown

MAY 1994

Transplantation

Editor: Lewis Burrows

Available now:

Festschrift for Rosalyn S. Yalow: Issues in

Hormones, Metabolism, and Society Medical Ethics

Eugene W. Straus, editor Daniel A Moros and Rosamond Rhodes, editors

volume 59, no. 2. March 1992 volume 60, no. 1, January 1993

96 pages SI 5 84 pages *15

Subscriptions for 1993: $65 indloiduals. US.; $70 all libraries: $75 indioiduals outside US.

To order subscriptions or copies of back issues ($15 each), please send check, paifable to The Mount SInal Journal of Medicine, to the Journal at Box 1094, One Gustaoe L Leoy
New York, NY 10029-6574: counUr sates at Suite IE, 50 E. 98th Street, New York, /VY (phone: 212 241-6108: FAX: 212 722-6386).

Toward the Conquest of Pain

Allan P. Reed, editor
volume 58. no 3, May 1991
84 pages tl5

#THE *

CM N(W YOtK

>10am" SINAI JOURNAL
DF MEDIQNE

OLUME 60 NUMBER 6

NOVEMBER 1993

c ::•

Education, Clinical Services, and
Research on Treating Older People:

The Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center — A Decade of Experience

Myron Miller, Rein Tideiksaar, and Carol Capello, guest editors

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center

Veterans

Affairs

Beth
Israel
Medical
Center

The Mount Sinai Journal of Medicine is published by The Mount Sinai Medical Center of
I 11 New York and has the following affiliates: Beth Israel Medical Center, New York; Bronx

MOUNT SINAI
JOURNAL OF
MEDICINE

Veterans Affairs Medical Center, New York; and Elmhurst
Hospital Center, New York.

Editor-in-Chief

Sherman Kupfer, M.D.

Editor Emeritus

Lester R. Tuchman, M.D.

Associate Editors

Harriet S. Gilbert, M.D. Julius Wolf, M.D.

Managing Editor

Claire Sotnick

Business and Production Assistant

Karen Schwartz

Assistant Editors

Stephen G. Baum. M.D.
David H. Bechhofer, Ph.D
Constanin A. Bona. M.D..
Edward J. Bottone. Ph.D.
Jurgen Brosius. Ph.D.
Lewis Burrows, M.D.
Joseph S. Eisenman. Ph.D.
Adrienne M. Fleckman, M.D
Richard A. Frieden. M.D.
Steven Fruchtman. M.D.
Paul L. Gilbert. M.D.
James H. Godbold, Ph.D.

Richard S. Haber, M.D.
Noam Harpaz, M.D.
Ph.D. Dennis P. Healy. Ph.D.
Tomas Heimann. M.D.
Barry W. Jaffm, M.D.
Andrew S. Kaplan. D.D.S.
Samuel Kenan. M.D.
Suzanne Carter Kramer, M.Sc.
Mark G. Lebwohl. M.D.
Kenneth Lieberman. M.D.
Charles Lockwood, M.D.

Lynda R. Mandell. M.D.. Ph.D.

Steven Markowitz, M.D.

Bernard Mehl. D.P.S.

Myron Miller. M.D.

Edward Raab, M.D.

Allan Reed. M.D.

Allan E. Rubenstein. M.D.

David B. Sachar. M.D.

Henry Sacks. M.D.

Robert Safirstein, M.D.

Ira Sanders, M.D.

Martin H. Savitz, M.D.
Clyde B. Schechter, M.D.
Michael Serby. M.D.
Phyllis Shaw, Ph.D.
George Silvay. M.D.
Barry D. Stimmel, M.D.
Nelson Stone, M.D.
Max Sung. M.D.
Carl Teplitz, M.D.
Rein Tideiksaar, Ph.D.
Richard P. Wedeen. M.D.

Editorial Board

Barry S. Coller. M.D.
Barry Freedman. M.B.A.
Richard Gorlin, M.D.

Nathan Kase. M.D. Jack G. Rabinowitz, M.D.

Panayotis G. Katsoyannis, Ph.D. John W. Rowe, M.D.
Charles K. McSherry, M.D. Alan L. Schiller, M.D.

Alan L. Silver. M.D.
Alvin S. Teirstein, M.D.
Rosalyn S. Yalow, Ph.D.

The Mount Sinai Journal of Medicine (ISSN No. 0027-2507; USPS 284-860) is published 6 times a year in January, March, May,
September. October, and November in one indexed volume by the Committee on Medical Education and Publications (Owner). The
Mount Sinai Hospital. Box 1094. One Gustave L. Levy Place. New York. NY 10029-6574. Subscription price: individuals. U.S.,
S65 per year; libraries, $70; individuals outside the U.S.. $75. Single copies. $15. New subscriptions begin with the first issue of
the calendar year. Send notice of change of address 60 days before the change is effective. Second-class postage paid at New York,
NY and at additional mailing offices. POSTMASTER: Send address changes to The Mount Sinai Journal of Medicine, Box 1094, One
Gustave L. Levy Place. New York, NY 10029-6574. Copyright 1993 The Mount Sinai Hospital.

THE

MOGNT SINAI JOURNAL
OF MEDICINE

Volume 60
Number 6
November 1993

CONTENTS

EDUCATION, CLINICAL SERVICES, AND RESEARCH ON TREAT-
ING OLDER PEOPLE:

The Department of Geriatrics and Adult Development, Mount
Sinai Medical Center — A Decade of Experience

Myron Miller, Rein Tideiksaar, and Carol Capello, editors

Introduction Robert N. Butler 455

EDUCATION

Education in Geriatrics: A Required Curriculum for Medical Stu-
dents Gabriel Gold 461

A Mount Sinai Student Reflects on His Experience in Geriatrics

Research Mathew Maurer 465

Geriatrics Fellowship Training at Mount Sinai Medical Center: A

Decade of Experience Monte H. Peterson 468

CLINICAL SERVICES

The Geriatric Evaluation and Treatment Unit: A Model Site for
Acute Care of the Frail Elderly, Education and Research

Howard Fillit and Myron Miller 475

The Phyllis and Lee Coffey Ambulatory Care Clinic: Ten Years of
Evolution in Treating Older Adults Barbara Paris, Diane E.
Meier, Jane Morris, Nurit Ginsberg, and Linda Weiss 482

Health Care for the Homebound Older Adult: A Medical Model

Laura Di Pollina, Gabriel Gold, and Diane E. Meier 488

Community Initiatives to Improve the Lives of Older Adults in

East Harlem Judith L. Howe 492

MANAGEMENT OF GERIATRIC DISORDERS

Practical Pharmacology for Older Patients: Avoiding Adverse

Drug Effects Judith Ahronheim 497

Evaluation and Management of Urinary Incontinence in Older

Patients Perry Starer 502

Falls in Older Persons Rein Tideiksaar 515

Falls Prevention: The Efficacy of a Bed Alarm System in an
Acute-Care Setting Rein Tideiksaar, Clifford F. Feiner, and
Jan Maby 522

continued

Volume 60
Number 6
November 1993

CONTENTS continued

Geropsychology and Neuropsychological Testing: Role in Eval-
uation and Treatment of Patients with Dementia Rajendra
Jutagir 528

Clinical Evaluation of Dementia Jonathan Koblenzer 532

Osteoporosis in the Aged Carolyn Murray and Diane E. Meier 539

The Principle of Autonomy and the Older Patient Diane E. Meier 546

LONG-TERM CARE

A Teaching Nursing Home: Ten Years of Partnership between
The Jewish Home and Hospital for Aged and the Mount Sinai
School of Medicine Leslie S. Libow 551

Ethical Issues in the Nursing Home Ellen Olsen 555

Nursing Staff Attitudes on the Use of Physical Restraints in a
Teaching Nursing Home Janet Michello, Richard R. Neu-
feld, Michael Mulvilhill, and Leslie S. Libow 560

Research Activities in the Department of Geriatrics and Adult
Development John H. Morrison, Jon Gordon, Myron Miller,
Howard Fillit, and Diane E. Meier 565

Public Policy and Geriatrics: The International Leadership Center
on Longevity and Society of the Department of Geriatrics
and Adult Development at Mount Sinai Medical Center

Malvin Schecter 574

List of Journal Reviewers in 1993 578

The index to Volume 60 will be published
in January 1994 (Volume 61, issue no. 1).

Index to Advertisers

Course on Myofascial Pain

Diagnosis and Treatment

page 579

Post-Graduate School of Medicine

Continuing Medical Education Courses

page 580

Introduction

Robert N. Butler, M.D.
Brookdale Professor and
Chairman

Department of Geriatrics and
Adult Development
Mount Sinai School of Medicine
The Mount Sinai Hospital

Education,
Clinical
Services,
And Research
On Treating
Older People:
A Decade of
Experience

The goal of the Department of Geriatrics and Adult Development
is to improve the quality of life of older people through excellence
in patient care, professional and public education, biomedical and
clinical research, and public policy development. The Department
strives to serve as a model and catalyst for academic geriatrics in
the United States.

Mount Sinai is a fitting home for this first and only Depart-
ment of Geriatrics in an American medical school since this insti-
tution has a unique history of contribution to the development of
geriatrics. In 1909 Ignatz L. Nascher, a Mount Sinai physician,
introduced the word geriatrics, and in 1914 he authored the first
American textbook of geriatrics. Alvin Goldfarb pioneered geriat-
ric psychiatry both at Mount Sinai and the Jewish Home and Hos-
pital for Aged (JHHA), where Frederick Zeman contributed to the
development of geriatric medicine and also wrote historical arti-
cles on geriatrics.

Mount Sinai continued its role as the pioneering home of ge-
riatrics with the establishment of the Department of Geriatrics
and Adult Development 10 years ago. Thomas Chalmers, then
president and dean, and Albert Stern, then chairman of the Board
of Trustees, agreed to create a full-scale department that could
claim curriculum time for geriatrics throughout the medical school
career, including a fourth-year required rotation, that would have
a relationship with a "teaching nursing home," and that would
include inpatient beds and an outpatient clinic. These goals, as
well as others, have been met in the department's first decade.

We worked to establish a science base in molecular biology,
neurobiology, and the immunology of aging. We were able to es-
tablish the endowed Mathers chair in molecular biology and the
Johnson chair in neurobiology and to place the professors in the
relevant centers at Mount Sinai to insure the strength that comes
from having a critical mass of collaborators as well as shared re-
sources. The Brookdale Foundation, which supported our tenth
anniversary celebratory symposium "Strategies to Delay Dysfunc-
tion in Later Life," has also supported an endowed professorship
and many of the department's efforts. We look forward to having
space in Mount Sinai School of Medicine's new research building,
where we expect to establish a major laboratory on aging and
longevity with special emphasis on neural aging. The central ner-
vous system — the mind — is the essential pacemaker of aging, both
in the biological sense and in terms of quality of life, since the
central nervous system and mental faculties are essential for per-
sonal adjustment and social adaptation.

The Greenwall chair is the first endowed chair for a professor-
ship shared between a long-term care institution, the Jewish Home
and Hospital for Aged (JHHA), and a medical school. We have
enjoyed an extraordinary partnership with the JHHA. With a
"bridge nurse," we have created opportunities to study the nature
of patient transfers from JHHA to Mount Sinai, as well as to insure
the best care.

Education and Clinical Services

As a full-scale department, we have carried out undergradu-
ate, postgraduate, and continuing medical education, provided
clinical services, and conducted research. In addition, we have un-

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

455

456

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

dertaken policy-related studies, pursued commu-
nity activities, and supported studies in bioethics.
We have developed special clinics devoted to the
top three concerns in geriatrics: dementia, prob-
lems of mobility, and incontinence. In coordina-
tion with the Department of Obstetrics and Gy-
necology, we established New York City's first
osteoporosis clinic, and with the Department of
Psychiatry, one of the National Institute on Ag-
ing Alzheimer's Disease Research Centers.

High-quality education in any branch of
medicine is impossible without the availability of
fine clinical services. Mount Sinai is, therefore,
fortunate to have two excellent sites at which stu-
dents see the principles of geriatrics at work. One
is the Geriatric Evaluation and Treatment Unit,
endowed by the Eisner Family Foundation and
housed in the Guggenheim Pavilion. The other is
the outpatient clinic, which was endowed by
Phyllis and Lee Coffey.

Medical teaching and learning depend not
only on first-rate clinical services but also on ex-
posure to the processes of scientific inquiry. In
scientific inquiry, the first activity is careful ob-
servation. The second is hypothesis-building,
wherein a proposition to be tested has an equal
chance of being refuted or confirmed. This hap-
pens in fine medical diagnostic inquiry as well as
in research. The department believes students at
all levels need to be exposed to these processes of
scientific inquiry, and it supports research elec-
tives by students and research by fellows.

Recruiting outstanding young people into the
field of geriatrics is vital, but it is extremely dif-
ficult. Nationally there are relatively few role
models in geriatrics. In general, interest in pri-
mary care medicine is declining, and physician
reimbursement for primary care and geriatrics is
not competitive. Training funds are minimal, and
there has not been a major national initiative to
support geriatrics comparable to the support of,
for example, cardiology. After the National Heart
Institute was established in 1947, some 16,000
cardiologists were trained during the first 22
years. Nothing like that occurred with the estab-
lishment of the National Institute on Aging, de-
spite our wishes.

There are, however, some efforts underway to
enhance education in geriatrics. The Brookdale
Foundation's National Fellowship Program,
which I chair, has helped create an outstanding
coterie of young people and moved them beyond
their fellowships into junior faculty positions in
geriatrics and in social and behavioral gerontol-
ogy. And the Commonwealth Fund, with the help
of the Alliance for Aging Research, is now en-

gaged in trying to develop a national physician-
scientist program in geriatrics.

I presented the concept of a federal geriatrics
initiative before the Senate Special Committee on
Veterans' Affairs, chaired by Senator John D.
Rockefeller IV, in May 1993. Funds drawn from
the Medicare graduate medical education pro-
gram and from Veterans Affairs would be coordi-
nated (without damage to existent residency pro-
grams) to foster the development of academic
geriatrics within each of the 126 medical schools
through core funding. Incentive payments would
go to those medical schools that develop addi-
tional specific programs, such as relationships
with teaching nursing homes, home care pro-
grams, a healthy elderly program, a hospice pro-
gram, mandated clerkships. We have made simi-
lar suggestions before the New York State
Legislature.

Also in the Department's education reper-
toire, our Healthy Elderly program at the 92nd
Street Young Men's- Young Women's Hebrew As-
sociation, with the inspired support of Janet
Fisher, has helped students overcome the nega-
tive stereotypes about old age that follow from
exposure only to sick and demented older persons.
Hunter College's Brookdale Center on Aging, su-
perbly directed by Rose Dobrof, and the depart-
ment have co-sponsored the federally supported
Geriatric Education Center, which helps educate
a wide variety of health and social service provid-
ers in geriatrics.

Community and
Policy Concerns

Our involvement in the East Harlem commu-
nity is perhaps best illustrated by our medical
home care program and by Project Linkage,
which is currently in development. The purpose of
Project Linkage is to create shared housing for
older persons in East Harlem, incorporating an
after-school enrichment program for latchkey
children between 5 and 12 years of age. The en-
richment program would include tutorials pro-
vided by retired older persons, which demon-
strates one way of using a still-productive
reservoir of great talent.

We established in 1990 the International
Leadership Center on Longevity and Society
(ILC) together with Japanese colleagues. The ILC
is the first private, nonpartisan, nonprofit inter-
national center on research, education, and policy
development devoted exclusively to the effects of
longevity on society and its institutions. The ILC
focuses not on individual aging, but on population

Vol. 60 No. 6

INTRODUCTION— BUTLER

457

aging and its impact. There are now two ILC of-
fices, one in Tokyo and one in New York City. We
hope soon to have a European center in Paris. The
ILC is based on the belief that it is possible to
undertake important comparative studies that
will yield information of value to collaborating
societies. We also believe that people who are not
part of a given society may be in a good position to
offer insights that are difficult for people within
the society to make. So far we have been collab-
orating with Japan on issues of the bedridden,
relationships among the generations, character-
istics of the nonprofit voluntary sector, and stud-
ies of older people living alone. The endowed
Harold Hatch International Lectureship in Geri-
atrics, part of the ILC, has brought us scholars
from Europe and Japan.

The Future of Geriatrics

Geriatrics does not need to become a practice
specialty. Rather, it should be primarily an aca-
demic and consultative specialty. An academic
specialty provides the leaders, the innovators,
and the researchers who can assure the integra-
tion and mainstreaming of knowledge about ag-
ing throughout all of primary care and specialty
medicine. Geriatrics should also be involved in
settings with high concentrations of older per-
sons, such as nursing homes and retirement com-
munities.

An unintended consequence of the diagnosis-
related-groups (DRGs) methodology, which was
introduced as a cost-containment strategy and
has led to a shorter length of stay in hospitals, is
that today it is no longer possible for a medical
student to learn about the natural history of med-
ical conditions in depth and to get to know pa-
tients and their families. To remedy this situa-
tion, I suggest that medical students follow the
patient through various sites. The department's
complex multisite educational program may be-
come a model for the development of a new-style
multisite (non-hospital-based) medical school of
the 21st century.

As I am writing this, we all eagerly await the
imminent presentation of the new health reform
package by the Clinton administration. One po-
tential problem that the administration needs to
guard against in the development of the reforms
is curtailing benefits that many people now have
while raising taxes. A single uniform payment
system is as important as a single payer method-
ology. In the Coffey Clinic, for an initial compre-
hensive evaluation of an older patient who has
multiple, complex, and interacting psychosocial
and physically acute and chronic problems, we

receive — depending on the payer — approximately
$9.50 from Medicaid, or $88 from Medicare, or an
average of $122 from other third-party payers. It
is not hard to see how unattractive poor and older
patients are in a free market economy. But if uni-
form payments were made for services and proce-
dures independent of the source of payment, such
discrimination would disappear.

Discrimination by disease is also a problem.
If one has a heart attack, for example, one is well
covered. Yet if one has a "brain attack," such as
dementia, one does not enjoy the same financial
protection. The first is treated in a hospital and
covered by Medicare; the second requires long-
term care, for which there is Medicaid coverage
only after "spending down" assets and income.

Obstacles in the development of geriatrics in-
clude not only reimbursement but the failure to
have an "intensity index": the older the patient,
the more complex the case and the longer both
diagnosis and care take. Great Britain, for exam-
ple, deals with this problem by adding to capita-
tion payments for older patients.

Although geriatrics does not have reimburs-
able technological procedures, it may soon have
Medicare payment codes for assessment and for
coordination of care. Moreover, Medicare may
soon become the template for general payments to
physicians, which would put all patients on the
same footing. The politics of all this is on the side
of geriatrics because of economic necessity and
demographic changes.

As for the future of geriatrics as it is taught
and practiced in the department, we have
achieved much, but there is always room for im-
provement. Through the JHHA, we have sur-
veyed the 12,000 students that have been trained
through our program and found that 70% appre-
ciated having been taught geriatrics at Mount
Sinai School of Medicine. The medical residents
have accepted their month-long rotation through
the program, but we regret that we are not able to
provide them a more comprehensive education,
including long-term care and other experiences,
as well as continuity in their contact with pa-
tients. We hope that with time these limitations
will be overcome. Of the 40 fellows who have
graduated from the program, some 70% have
filled academic positions in geriatrics. We are
pleased that this percentage reflects the mission
of the department.

We remain concerned about the central chal-
lenge of geriatrics — the maintenance and restora-
tion of function in our older patients. We must
deal with the issue more effectively in both our
inpatient unit and outpatient clinic. With hospi-

458

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

talization comes a loss of function, and we are
endeavoring to make the hospital experience less
traumatic for the older patient. Another great
need is to study and improve the conditions for
older persons in American emergency rooms.

In this century we have increased average
life expectancy at birth by 28 years in the United
States. The Baby Boomers, the largest generation
in U.S. history, which comprises one-third of our
nation's population, will become a huge cohort of
older persons between the years 2020 and 2030.
In the judgment of most people in gerontology and
geriatrics, our society is ill-prepared to meet the
challenges that this increase will bring. There-
fore, the Baby Boomer generation is very much at
risk, as are the generations that will follow.

Despite advances in the field of aging, people
in general continue to feel discomfort about the
topic. This unease translates into denial and
avoidance. The sense of immortality of adoles-
cents is legendary; there is a comparable legend-

ary lifelong avoidance of the intimations of aging
and mortality with which we all must deal. Mem-
bers of the department continue to confront the
issues of ageism, as well as our societal and cul-
tural taboos with regard to older people, includ-
ing their sexuality, dying, and death.

The department's work has not been easy,
and our job is not over. Because changes are im-
minent in both demographics and health care de-
livery in the United States, the future is some-
what uncertain. What is certain, however, is that
geriatrics as a medical field will grow in impor-
tance and should be at the center of discussion. I
hope this issue of The Mount Sinai Journal, "Ed-
ucation, Clinical Services, and Research on Treat-
ing Older People: The Department of Geriatrics
and Adult Development, The Mount Sinai Medi-
cal Center — A Decade of Experience," which de-
tails the innovative programs and the accom-
plishments of the department, will contribute to
those discussions and to their resolution.

Education, Clinical Services, and Research
On Treating Older People

Education

Education in Geriatrics:

A Required Curriculum for Medical Students

Gabriel Gold, M.D.
Abstract

The mandatory geriatrics and gerontology curriculum at The Mount Sinai School of Med-
icine in New York includes two modules for first- and second-year students and a four-
week block experience for fourth-year students. The first-year curriculum emphasizes
socioeconomic, psychosocial, biomedical, and attitudinal issues. The second-year experi-
ence serves as an introduction to clinical geriatrics. The fourth-year clerkship allows
students to further develop their fund of geriatric knowledge, learn specific geriatric
skills, and build on their internal medicine foundation, integrating new knowledge and
skills and developing into comprehensive practitioners who can apply the team approach
to address all the medical, functional, psychosocial, and ethical aspects of caring for the
elderly.

Current changes in demographics include a
dramatic increase in the elderly population in the
United States, particularly in the group aged over
85 years. The proportion of individuals over age
65 rose from 4% of the total population in 1900 to
11% in 1980, and it is projected to reach 20% by
year 2030 (1). The number of individuals over age
85 rose from 370,000 in 1940 to 2.3 million in
1980, and it is estimated to grow to 12.8 million in
2040 (2). Furthermore, the elderly represent a
disproportionately large part of all medical con-
sumers. Individuals over age 65 use hospitals at
2.8 times the rate of those under 65 (3) and ac-
count for 29% of all health care expenditures in
the United States (4). The over-85 age group uses
hospitals tw^ice as much as those aged 65 to 69,
and nursing homes 11 times as much for women
and 16 times as much for men (5). It is predicted
that by the year 2020, the elderly will consume
nearly half of the nation's health care services (6).
The challenge of providing health care to this

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, New York, NY. Address
reprint requests to Gabriel Gold, M.D., Department of Geri-
atrics and Adult Development, The Mount Sinai Medical Cen-
ter, One Gustave L. Levy Place, Box 1070, New York, NY
10029.

ever-increasing segment of the population has
heightened the need to teach geriatrics in medical
schools. In 1978 the Institute of Medicine recom-
mended that appropriate content on aging be in-
cluded in basic and clinical science courses and
favored a required course that integrates knowl-
edge about aging and the problems of the elderly
(7). In 1983, the Association of American Medical
Colleges departed from its traditional reticence in
addressing categorical subjects in medical school
curricula and published a guideline for geriatrics
curriculum assessment, outlining ways in which
courses in basic science and clinical medicine
could be strengthened in the areas of geriatrics
and gerontology (8). Although many medical
schools now offer training in geriatrics and ger-
ontology, the structure of these programs varies
greatly in time and content; in addition, some are
mandatory, others are selective, and still others
are elective. Initiatives of the Department of Ge-
riatrics and Adult Development at the Mount
Sinai School of Medicine have led to both elective
and mandatory undergraduate training in geriat-
rics.

Elective training includes research and clin-
ical electives in geriatrics for first- and second-
year students, as well as a Geriatrics Student In-
terest Group for those who wish to cultivate their

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

461

462

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

interest in different aspects of the aging process.
Mandatory training is described below.

Geriatrics Module in the
Human Development Course

Thirteen of the total 63 hours of the Human
Development course for first-year students are
devoted to geriatrics. The goals of this module are
to increase the student's knowledge of socioeco-
nomic, psychosocial, and biomedical issues re-
lated to aging. Attitudinal issues are also stressed.

Lectures in the geriatric module are:

Introduction to

Geriatrics
Aging Physiology
Theories of Aging
Aging Psychology
Mental Health and

Aging

Osteoporosis
Mobility: Gait and

Balance
Sexuality and Aging
Sociology of Aging
Ethical Issues

In addition to lectures, students attend a two-
hour small group workshop, facilitated by geriat-
rics faculty, in which three to four older persons,
representative of the positive aspects of aging,
discuss their personal aging experiences and
share their wisdom and thoughts on aging with
students. The primary goal of these small groups
is to help students address early in their medical
careers any stereotypes they may have concern-
ing the elderly; they see first-hand that aging can
be healthy and productive. Workshops are fol-
lowed by an optional reception, where interested
students and elders mingle and continue their
earlier discussions more informally.

Geriatrics Module in the
Physical Diagnosis Course

Three hours of the second-year Physical Di-
agnosis course are devoted to the history-taking,
physical examination, and functional assessment
of the geriatric patient. Students are alerted to
the major differences between obtaining histories
and performing physical examinations on older
versus younger people. Several points are
stressed, such as the atypical presentation and
multiplicity of disease and the need to examine
patients for bedsores, to assess their mobility and
functional and social status, and to perform a
mental status examination. After a lecture, stu-
dents break into small groups to observe a geria-
trician taking an elderly patient's complete his-
tory.

The Geriatrics Clerkship

A required four-week geriatrics clerkship for
all fourth-year students emphasizes attitudinal
aspects, clinical skills, and clinical knowledge of
common geriatric syndromes. The objectives of
the clerkship were developed based on the medi-
cal literature (8-10) and the polling of all geriat-
rics faculty at a department retreat. Students ro-
tate through one of six sites and are exposed to
multiple levels of geriatric care, which may in-
clude acute geriatric care and geriatric assess-
ment; geriatric consultation; long-term care; out-
patient geriatric medicine; home care; hospice
care; and well-elderly programs. The intensity of
exposure to the varied aspects of geriatric medi-
cine is dependent on the site.

The Mount Sinai Hospital is a large tertiary-
care university hospital, which includes a 16-bed
Geriatrics Evaluation and Treatment Unit that
provides acute medical care and geriatric assess-
ment to frail elderly persons. Students rotating
through the unit are assigned two to four patients
and are responsible for preparing a written his-
tory, physical, and geriatric assessment for these
patients. Geriatric assessment may include func-
tional assessment; gait and falls assessment; cog-
nitive assessment; dental, hearing, and vision
assessments; nutritional assessment; health
maintenance; evaluation of skin integrity; risk of
pressure sores and need for preventive measures
and treatment; need for rehabilitation services;
evaluation of urinary or fecal incontinence; as-
sessment for possible elder abuse and caregiver
stress; and polypharmacy assessment. Students
also spend one half-day each week in the outpa-
tient geriatric clinic, accompany a physician on
home visits, and visit the Rehabilitation Depart-
ment, a long-term-care facility, and a hospice.

Students assigned to the Geriatrics Consul-
tation and Liaison Service at Mount Sinai are re-
sponsible for presentation of all new consult
cases, which includes a geriatric assessment as
described above. These students spend one half-
day each week in an outpatient setting (outpa-
tient clinic or private-practice office) and one day
each week accompanying a hospice physician on
home visits. These students also visit the Rehabil-
itation Department and a long-term-care facility.

Students assigned to the City Hospital Cen-
ter at Elmhurst, a 780-bed municipal teaching
hospital, are responsible for geriatric consulta-
tion on frail elderly patients on acute care wards.
Students participate in weekly rehabilitation
clinics and spend one half-day each week in the
geriatric outpatient clinic and one day each week

Vol. 60 No. 6

EDUCATION— GOLD

463

at a well-elderly program (11) that has had a pos-
itive impact on student attitudes toward the el-
derly (12). Students also visit a long-term-care
facility.

The Jewish Home and Hospital for Aged has
two campuses that serve as clerkship sites. Stu-
dents at the Central House campus rotate
through a 514-bed nursing facility; they are as-
signed both acutely and chronically ill patients
and are responsible for an in-depth geriatric con-
sultation on all medical, ethical, functional, and
psychosocial aspects of each case. Students also
spend at least one full day on home visits and
attend rehabilitation clinic for two half-days. Stu-
dents at the Kingsbridge Division campus rotate
through an 816-bed nursing facility, a 300-tenant
enriched housing facility, and a geriatric day-care
program; they are responsible for geriatric con-
sultation on rehabilitation patients and acutely
ill nursing facility residents. Students also spend
one day a week in specialty clinics for nursing
facility residents, one day in an outpatient clinic,
and one to two days accompanying a physician on
home visits.

Students rotating at the Bronx Veterans Af-
fairs Medical Center are assigned to a 120-bed
nursing facility or to a 25-bed intermediate-care
facility and a 15-bed geriatric evaluation and
management unit. Students assume direct re-
sponsibility for the care of four to six residents
and also spend two half-days in a geriatric home-
care program.

At each of these sites, students participate
in geriatric medicine rounds, geropsychiatry
rounds, specialty rounds, and an ethics teaching
session. Teaching methods include lectures, small
group patient-management sessions, individual
student oral presentations to their peers, a sylla-
bus, a simulated patient encounter, and informal
teaching during clinical activities.

Two days of core lectures are given at Mount
Sinai School of Medicine during the first week of
the clerkship, covering:

The syllabus supports the core lecture series and
is also designed for self-study. Each chapter is
preceded by a list of specific learning objectives
and followed by a series of test questions, clarify-
ing for students what they are expected to learn
and allowing them to test their knowledge once
they have read the material.

Patient-management sessions consist of
small-group precepted workshops using a paper
case that often emphasizes specific aspects of ge-
riatric medicine, stresses diagnostic reasoning
and therapeutic decision making, and provides an
opportunity for students to integrate and apply
their newly acquired skills and fund of knowledge
to real-life cases (box).

Ethics teaching includes informal bedside
discussions, ethics rounds, and a half-day session
with a faculty preceptor at which medical stu-
dents are brought together, watch a videotape
presentation of an ethical dilemma, discuss their
opinions about the case, and relate them to basic

Introduction to

Geriatrics
Osteoporosis
Cognitive Impairment
Falls and Immobility
Medical Insurance and
Support Programs
For the Elderly

Demographics
Health Care Costs
How Doctors Get Paid
Urinary Incontinence

Rehabilitation in the

Elderly
Nutrition
Restraints in the

Nursing Home
Fecal Incontinence
Geriatric Dentistry and

Oral Pathology

Infections in the
Nursing Home
Tuberculosis
Thyroid Disease
Constipation
Sexual Dysfunction
Health Maintenance

Patient Management Session Topics

Falls

Urinary incontinence

Fecal incontinence

Dementia

Depression

Elder abuse

Ethics

Thyroid disease
Diabetes mellitus
Malnutrition
Peripheral vascular

disease
Acute abdomen
Acute cholecystitis
Peptic ulcer disease
Atrial fibrillation
Sick sinus syndrome
Syncope
Osteoarthritis
The older driver

Hip fracture

Delirium

Acute parotitis

Anemia

Polypharmacy

Nonsteroidal

anti-inflammatory

drugs
Digoxin toxicity
Asymptomatic carotid

bruit
Hypertension
Pressure sores
Health maintenance
Stroke

Immunity and

infections
Water and electrolyte

disturbances
Atypical presentation

of disease

Subsequent lectures on the following topics are
given at each site; the topics may vary depending
on clerkship site and rotation:

464

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

ethical principles. This session is generally well
received by students, who view it as a significant
improvement over the lecture format previously
used.

In the 1991-92 academic year, an innovative
teaching method was implemented, using "stan-
dardized patients" (SPs) — individuals trained to
portray real patients in a reproducible and stan-
dardized fashion. Approximately half of all
fourth-year students participate in this new pro-
gram, in which they take a focused history from
or a physical examination of an SP in an exam-
ining room equipped with TV cameras and micro-
phones, and answer a postencounter question-
naire addressing planned diagnostic studies and
differential diagnosis. After the encounter, each
SP completes a checklist of history and physical
items performed, and a faculty preceptor, who
watched the interaction on a TV monitor, pro-
vides feedback to students and hands them a tape
of their encounter for review. The vast majority of
participating students rated this new program
positively.

Student grades (pass, fail, honors) are based
on clinical performance and oral presentation. In
addition, students must pass a final multiple
choice, minimum competency, written examina-
tion.

The geriatrics clerkship has been shown to
result in significant knowledge gain and has been
positively evaluated by the vast majority of stu-
dents (13).

Four off-campus sites are also available for a
limited number of students, at the Instituts Uni-
versitaires de Geriatrie de Geneve and the World
Health Organization in Geneva, Switzerland, the
National Institute of Aging's Gerontology Re-
search Center and Johns Hopkins' Francis Scott
Key Hospital in Baltimore, the Hopital Broca's

geriatric facility in Paris, France (University of
Paris), and the Soroka Medical Center's geriatric
facility in Beer Sheva, Israel (Ben Gurion Uni-
versity).

References

1. Brody J A. Facts, projections, and gaps concerning data on

aging. Public Health Rep 1984; 99:468-^75.

2. Rosenwaike I. A demographic portrait of the oldest old.

Milbank M Fund Q 1985; 63:187-205.

3. The National Task Force on Gerontology and Geriatric

Care Education in Allied Health. Part I. The implica-
tions of an aging society for health care needs. J Allied
Health 1987; 16:307-322.

4. Kane R, Solomon D, Beck J, et al. The future need for

geriatric manpower in the United States. N Engl J Med
1980; 302:1327-1332.

5. Soldo BJ, Manton KC. Health status and service needs of

the oldest old: current patterns and future trends. Mil-
bank M Fund Q 1985; 63:286-319.

6. AMA Council of Scientific Affairs. Societal effects and

other factors affecting health care for the elderly. Arch
Int Med 1990; 150:1184-1189.

7. Institute of Medicine. Aging and medical education.

Washington, D.C.: National Academy of Sciences, 1978.

8. Executive Council of the Association of American Medical

Colleges. Preparation in undergraduate medical educa-
tion for improved geriatric care — a guideline for curric-
ulum assessment. J Med Educ 1983; 58:501-526.

9. Robbins AS, Fink A, Kosecoff J, et al. Studies in geriatric

education: I. Developing educational objectives. J Am
Geriatr Soc 1982; 30:281-288.

10. Dans PE, Kerr MR. Gerontology and geriatrics in medical

education. N Engl J Med 1979; 300:228-232.

11. Adelman R, Hainer J, Butler RN, Chalmers M. A well

elderly program: an intergenerational model in medical
education. Gerontologist 1988; 28:409-413.

12. Adelman RD, Fields SD, Jutagir R. Geriatric education.

Part II: The effect of a well elderly program on medical
student attitudes toward geriatric patients. J Am Geri-
atr Soc 1992; 40:970-973.

13. Fields SD, Jutagir R, Adelman R, et al. Geriatric educa-

tion. Part I: Efficacy of a mandatory clinical rotation for
fourth year medical students. J Am Geriatr Soc 1992;
40:964-969.

The Department of Geriatrics and Adult Development offers a great variety of clinical and
research-oriented experiences for students through its electives that supplement the required
geriatrics curriculum. Mathew Maurer, M.D., was one Mount Sinai medical student who took
advantage of these opportunities. — Myron Miller, M.D.

A Mount Sinai Student Reflects on His
Experience in Geriatrics Research

Mathew Maurer, M.D.

My fascination with the elderly began as a child
and has been further cultivated throughout my
medical education by formal coursework, re-
search opportunities, and clinical experiences. All
this has led to my decisions on my future career
direction.

Both of my grandparents always took an ac-
tive role in their nine grandchildren's lives — cel-
ebrating each birthday, each graduation, each
bar or bat mitzvah, and each wedding. It was my
relationship with my grandfather, Harry, how-
ever, which had the strongest impact on my life.

Harry continuously worked hard to prove his
vigor. On my seventh birthday, when my parents
presented me with a bicycle, Harry insisted on
taking it out for a test drive — even though he
walked with a cane. I will never forget how he
rode, his body swaying to and fro like a pendu-
lum, maintaining his balance only by constantly
turning the front wheel right and then left. I was
sure he would fall. Luckily, he didn't.

When my grandmother died, Harry was dev-
astated and became extremely isolated. Like most
men, he'd just assumed that she would outlive
him. My grandmother had been the one who had
done all the social planning, the one Harry had
always relied on to draw him out of his shell. I'm
not sure if he really liked himself after her death.
He sat in front of the television all day, kept a
poor diet, and rarely bathed.

The entire situation surrounding my grand-
father frustrated me, and I would not permit my-
self to communicate effectively with him in fear of

The author is a resident in internal medicine at Columbia
Presbyterian Hospital, New York City. Address reprint re-
quests to Myron Miller, M.D., at the Department of Geriatrics
and Adult Development, Box 1070, Mount Sinai Medical Cen-
ter, One Gustave L. Levy Place, New York, NY 10029.

what I might learn. To me, it seemed as if Harry
wanted to die, which I considered absurd and ir-
rational. I had been so personally invested in his
life that I could not let myself pursue the mat-
ter— I was afraid he would actually admit he
wanted to die.

After a while my parents sought medical at-
tention for Harry. Unlike anyone in my family,
the physician was in a position to ask the ques-
tions that needed asking. More important, Harry
was able to confess to him what he could not ad-
mit to his own family — that he wanted to die —
and the physician, naturally more objective than
Harry's family would be, was not shocked or hurt
by this confession. Harry admitted that since he
had spent his entire life with his wife, he felt his
days now had no order and his life no direction.

It was through this interaction of grandfa-
ther and physician that I began to realize the re-
markable position a physician is in both socially
and psychologically. He can pose questions that
need asking — questions that neither family mem-
bers nor close friends can or will ask — without
making the patient feel as if his privacy is being
invaded, because the impetus behind the ques-
tioning is benevolent, motivated by a desire to
assist, to improve, and, if necessary, to rescue.
The patient can answer honestly because any psy-
chological ramifications of his or her answers are
minimized, knowing that this professional is ob-
jective, has heard similar views before, and is
trained to address them.

Harry's admission of his feelings was thera-
peutic, because after his talk with his physician,
he began to improve. He would get up early each
morning, take a walk to buy the paper, and read
in our backyard. Once I woke up, he would join
me for breakfast and see me off to school. He
never failed to let me know how proud he was of

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

465

466

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

my accomplishments. He was happy to share once
again in another's life. We once again had a great
relationship, and I will always fondly remember
him.

During my second year of medical school, my
interest in geriatrics and gerontology was again
sparked when I was introduced to the psycholog-
ical aspects of illness in a Behavioral Science
course. Here, I learned more about my own expe-
riences with Harry. I came to appreciate that
physical illness occurs in a social and psycholog-
ical context that is unique to each patient, a con-
text whose influences, at times, play a more im-
portant role than the biological problem itself
Reading extensively about the impact of social
isolation on the health and well-being of elderly
patients, I soon recognized what a close correla-
tion there is between morbidity and mortality
and social isolation.

During the summer of my third year in med-
ical school, I was able to explore the impact of
social isolation in the elderly when I received a
competitive research grant issued by the John A.
Hartford Foundation, which encourages future
physicians to explore their interest in gerontol-
ogy. Since the cornerstone of geriatric medicine is
assessment and no tools to evaluate social isola-
tion had yet been validated, I developed a short,
self-administered questionnaire that could be
used in a primary care geriatric setting to iden-
tify patients in whom social isolation is an asso-
ciated factor in the development and perpetuation
of their illness.

Though excited that I could pursue a re-
search initiative, I was also frightened. Previ-
ously, I had considered research to be a pursuit
that narrowed one's perspective and, although I
supported research activities, I felt that success-
ful researchers were so focused on their particular
area of interest that they did not see the big pic-
ture. I also viewed researchers as loners who did
not want to interact with others — totally in con-
trast to the type of physician I wished to be: well-
rounded, compassionate, and intimately involved
with my patients.

However, by using my personal interest and
convictions to develop a hypothesis and explore it
scientifically, I realized that my view of a success-
ful investigator was incorrect; in truth, research
was a natural extension of the person I wanted to
be. In its inception, a hypothetical evaluation us-
ing the scientific method is narrow, focused, and
somewhat constraining — similar to how sand in
an hourglass meets at the center. But truly great
researchers do not stop there. After proving or
disproving their hypotheses, they begin to evalu-

ate how their conclusions relate to other fields in
medicine, psychology, and sociology. Now that
sand in the hourglass begins to fall through its
narrow focal point, mingling with all the other
sand particles, all the previous proven or dis-
proved hypotheses. The "inadequate" researcher
is stuck in that narrow focal point, but the suc-
cessful researcher moves on, broad-mindedly cou-
pling his or her field of interest with that of oth-
ers.

By the middle of my third year of medical
school, my initial research was completed, show-
ing

• A poor clinical consensus among health care
providers for evaluating a patient for social iso-
lation

• More frequent agreement among social work-
ers and nurses than any other combination of
health care team members

• Seven factors that are good predictors of social
isolation

— marital status
— pets

— frequency of familial or companion contact
— personal evaluation of adequacy of familial

or companion contact
— amount of television watched
— belief that he or she had no one with whom to

discuss problems
— belief that if he or she died, they would not be

missed by someone in their neighborhood

With the John A. Hartford Foundation pro-
viding financial support, I was able to travel to
Chicago to present my research findings at the
annual meeting of the American Geriatrics Soci-
ety/American Federation for Aging Research.

Yet, while my commitment to care for the
elderly was still strong, my research endeavor left
me somewhat unfulfilled. First, the research de-
sign did not provide statistically significant re-
sults, nor was it ideal for the objective I had in
mind. Second, the research was in gerontology
and not geriatrics. In light of my medically based
training, I believed that I could do more justice by
pursuing a biologically based hypothesis.

I was able to do so during my final year of
medical school. I fulfilled my fourth-year geriat-
rics rotation requirement in Baltimore, Md, by
volunteering for a liaison program between
Mount Sinai Medical Center and the National In-
stitute on Aging (NIA). I had already decided to
pursue a career in internal medicine and felt that
additional experience in one of the medical sub-
specialties would be beneficial. On joint recom-
mendations from the geriatrics faculty at Mount

Vol. 60 No. 6

A STUDENT'S EXPERIENCE— MAURER

467

Sinai and the administrative staff of the NIA's
Baltimore Longitudinal Study, I contacted Dr.
Jerome Fleg, a staff cardiologist at the Gerontol-
ogy Research Center of the NIA. Once I had read
numerous articles written by Dr. Fleg and his co-
workers, he and I agreed on a research area ab-
sent from the current literature, an area the NIA
had explored but never completed.

Through extensive data analyses, using the
database of the Baltimore Longitudinal Study of
Aging, a major NIA intramural research initia-
tive, I evaluated the prevalence and prognostic
significance of supraventricular tachycardia
(SVT) during exercise treadmill testing in a pro-
posed healthy population. After completing this
epidemiological survey in 1,443 participants, I
discovered the following:

• The prevalence of SVT during maximal exer-
cise treadmill testing significantly increases
with age in both men and women

• Exercise-induced SVT is not associated with
major coronary risk factors or exercise-induced
myocardial ischemia

• Episodes of SVT during maximal treadmill
testing usually are limited to short asymptom-
atic runs and predominantly occur in older sub-
jects

• Although the presence of exercise-induced SVT
does not portend a long-term increase in either
cardiovascular morbidity or mortality, individ-
uals do demonstrate a somewhat higher rate
pressure product at maximal effort than control
subjects, suggesting that left atrial pressure
may also be higher

• Most important, exercise-induced SVT, previ-
ously thought to have no prognostic signifi-
cance, may be a marker for future SVT on rest
electrocardiograms.

I presented these findings in Dallas at the Amer-
ican College of Cardiology's annual meeting.

I truly believe that geriatrics offers exciting
research opportunities. As people are living
longer, medicine is discovering that it needs to
extend its database to elucidate the normal phys-
iology for octogenarians and octogenarians-plus.
Scientists are just now beginning to define what
constitutes normal aging so we can differentiate
pathology from senescence. For example, addi-
tional research conducted at the NIA and at
Johns Hopkins indicates that an aging heart free
from cardiovascular disease maintains normal
cardiac output at lower heart rates by increasing
the end diastolic volume. (The aging heart relies
more on the Frank Starling mechanism than ad-
renergic response to maintain cardiac output.)
The clinical correlates to this are the elderly pa-
tients with myocardial infarction who have dys-
pnea, as opposed to the chest pain that occurs in
their younger counterparts. Thus, we do have a
clear example of how normal aging physiology
contributes to the altered presentation of disease
in elderly patients.

Through my own research activities, I hope
to contribute to the body of medical knowledge on
aging and, in turn, utilize this information to for-
mulate an understanding of age differences in
disease. In this way, I hope to better enable my
patients to achieve successful longevity. Looking
into the future, I see myself in academic medi-
cine, actively involved in geriatrics through var-
ious clinical, teaching, and research activities.

My experiences in geriatrics have been ex-
tremely fulfilling, both intellectually and so-
cially. I wholeheartedly encourage others to ex-
plore this vast option available in the new
medical frontier of geriatric medicine.

Geriatrics Fellowship Training at Mount

Sinai Medical Center:

A Decade of Experience

Monte H. Peterson, M.D.

The tenth anniversary of the Department of
Geriatrics and Adult Development at Mount
Sinai Medical Center provides an opportunity to
review chronologically the history of its fellow-
ship program in the context of the national evo-
lution of geriatric medicine and graduate train-
ing programs. This article updates an earlier
report (1) and provides some data on the outcome
of the trainees who have graduated from the pro-
gram.

Physician Personnel Needs in Geriatrics

An important stimulus behind the develop-
ment of fellowship programs in geriatric medicine
across the country has been the need for geriat-
rics personnel. A number of studies and reports
have addressed this issue (2-7), and Table 1 sum-
marizes some of their results. The recommenda-
tions of the various studies are dependent on the
validity of the underlying assumptions and meth-
odologies involved. The most recent estimates
come from Reuben and colleagues (6, 7).

The first of the most recent studies (6) deals
with the need for physicians (nonsurgical physi-
cians and geriatricians) to provide medical care to
the elderly. Three different economic scenarios
were considered: moderate growth, recession
economy, and steady growth. The projected need
for internist and family-physician geriatricians
for the year 2000 ranged from 3,668 (recession
economy) to 9,705 (steady growth). This was pred-

From the Department of Geriatrics and Adult Development,
Mount Sinai Medical Center, New York, NY. Address reprint
requests to Monte Peterson, M.D., Box 1070, Mount Sinai
Medical Center, One Gustave L. Levy Place, New York, NY
1002S-6574.

icated on an estimation of 43% of a geriatrician's
time being involved in clinical care of older per-
sons.

The other recent study (7) addresses the need
for geriatrics faculty. Five specialties (internal
medicine, family practice, neurology, psychiatry,
and physical medicine) were considered, and
within each specialty, estimates were made for
both physician and nonphysician faculty. Two
methods were employed, the first relying on an
advisory panel's estimates of the minimum fac-
ulty needed to sustain a division, and the second
method utilizing the judgment of members of the
American Geriatrics Society Education Commit-
tee on the number of faculty necessary to provide
a minimum core of geriatrics training for medical
students and residents in the five specialties. For
comparison. Table 1 gives the projected number of
physician faculty in internal medicine and family
practice needed to provide core training.

Fellowship Programs
In Geriatrics

The nation's first geriatrics fellowship pro-
gram was established in 1972 (8). As one indica-
tion of the growth of the field in the 21 years since
then, the upcoming sixth edition of the Directory
of Fellowship Programs in Geriatric Medicine,
published by the American Geriatrics Society,
will list approximately 150 programs (9). How-
ever, this is likely to be an inflated number, since
a review of the fifth edition (10) shows that this
number includes both fellowship and residency
training programs (that is, geriatrics training
within internal medicine and family medicine
residency programs). In addition, the listing in-
cludes anticipated as well as actual programs and

468

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

FELLOWSHIP TRAINING— PETERSON

469

TABLE 1

Estimates of Personnel Needs in Geriatrics

Year

Study
(ref. no.)

Estimated need

Time
frame

1980

1981

1987

1993

RAND/UCLA
(2)

RAND/UCLA

(3)

Institute of

Medicine

Committee on

Leadership

for Academic

Geriatric

Medicine (4)
Department of

Health and

Human

Services (5)
RAND/UCLA

(6)

RAND/UCLA
(7)

7000-10,000
geriatricians
(academic and
practicing)

900-1600 geriatric
medicine faculty

2100 faculty
geriatricians

1990

2000
2000

1000-2000 academic
geriatricians

3,668-9,705 internist
and family physician
geriatricians to
provide medical care,
depending on the
economic scenario

Physician faculty
needed to provide
core geriatrics
training: internal
medicine, 2320;
family practice, 1874

2000

Current

programs in Canada as well as the United States.
The degree to which fellowship programs are, in
fact, meeting the need for geriatrics personnel
(especially the need for academic geriatricians)
was addressed by Reuben et al., who estimated
that, due to attrition, only "25 percent of those
who are [fellowship] trained to be geriatrics fac-
ulty will actually achieve senior faculty status"
(7). They further point out that even if fellowship
positions increased in number by 50%, attrition
would result in a net loss of geriatrics faculty. The
optimal way to meet personnel needs in geriatrics
is currently a topic of much debate within the
field (11, 12).

Mount Sinai Fellowship
Program — Initial Formulation

Eight fellows began their training in geriat-
rics at The Mount Sinai Medical Center (MSMC)
in July 1983. It is helpful to conceptualize the
clinical training of the geriatrics fellowship pro-
gram as having block and longitudinal compo-
I nents (Table 2). In addition to Mount Sinai, the
initial clinical training sites were Mount Sinai's
teaching nursing home, The Jewish Home and

Hospital for Aged (JHHA), and the Elmhurst
Hospital Center.

Block Components

Geriatrics Inpatient Unit. Two fellows were
based in the department's fledgling Geriatrics In-
patient Unit, where they provided primary care
to the patients, as well as geriatric assessment
and management.

Geriatrics Consultation Service. Through its
activities, the Geriatrics Consultation Service
soon established itself as the department's most
wide-ranging contact with the rest of the hospital.
In addition to responding to requests for consul-
tation, fellows evaluated patients in the emer-
gency room for possible admission and followed
up patients known to the department who were
hospitalized on units other than the Geriatrics
Unit (for instance, a Geriatrics Clinic patient hos-
pitalized on the surgical service).

Geropsychiatry. The fellow on the geropsy-
chiatry rotation assumed primary care responsi-
bility for a cohort of four to five patients on the
inpatient Geropsychiatry Unit. The focus of the

TABLE 2

Mount Sinai Medical Center Geriatrics Fellowships Block
and Longitudinal Components, 1983 and 1993

Block
rotation

Longitudinal
component

1983

Year 1

unspecified

Year 2

1993
Year 1

Year 2

Geriatrics Inpatient
Unit

Consultation Service
Geropsychiatry
Skilled Nursing Unit/

Rehabilitation Unit,

JHHA
City Hospital Center

at Elmhurst
Research

Consultation and
Liaison Service

Group Practice,
JHHA

Rehabilitation,
JHHA

Bronx VAMC

Research I

Geriatrics Evaluation
and Treatment Unit
Geropsychiatry
Special Emphasis
Research II

Geriatrics Clinic

Coffey Geriatrics

Clinic
Longitudinal

model, JHHA

At Mount Sinai Medical Center unless otherwise specified:
JHHA, The Jewish Home and Hospital for Aged; VAMC, Vet-
erans Affairs Medical Center.

470

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

rotation was geriatric psychopharmacology and
the "medical-psychiatric interface," such as the
management of a depressed patient with cardiac
conduction disease.

Skilled Nursing and Rehabilitation Units,
JHHA. Most fellows received their first introduc-
tion to nursing homes and nursing home care at
The Jewish Home and Hospital for Aged. Three of
the skilled nursing units at the Manhattan Divi-
sion of JHHA were designated teaching units. A
rehabilitation unit was created, utilizing approx-
imately half of the beds on one of the units. Dur-
ing the first year of the fellowship, each fellow on
this rotation had primary care responsibility for
residents at both levels of care (skilled nursing
and rehabilitation).

Elmhurst Hospital Center, Queens. At Elm-
hurst, the site of the first geriatrics fellowship
program in the United States, the activities of the
fellow centered on the 72-bed skilled nursing unit
existing within the hospital. The fellow on this
rotation worked closely with geriatrics nurse
practitioners, both on the Skilled Nursing Unit
and in the Geriatrics Clinic.

Research. From its outset, the fellowship
program was designed with six months of pro-
tected time for research during the second year.

Longitudinal Component

Coffey Geriatrics Clinic. Ongoing clinical
experience and training was provided in the Phyl-
lis and Lee Coffey Geriatrics Clinic, where fellows
worked one or two half-days per week as part of
an interdisciplinary team.

Developments Since 1984

- Major changes in the fellowship program
have since evolved.

1984-1985. During the academic year 1984-
85, two didactic research seminar series were ini-
tiated. The objective of "Human Aging Research"
was to give fellows an appreciation of both the
generic issues in aging research (such as funding)
and the specific topics currently being pursued by
department faculty. This was followed by a course
in research methods that focused on epidemiol-
ogy, research design, and biostatistics. Increased
time was also allotted to research by creating a
protected research rotation during the first year
at JHHA.

1986-1987. In the academic year 1986-87, no-
table changes occurred in rotations involving
Mount Sinai affiliates.

At The Jewish Home and Hospital for Aged,
a second longitudinal clinical component was

added to the program by implementing the "lon-
gitudinal model," in which each fellow was as-
signed patient-care responsibility for a small co-
hort of residents on one of two skilled nursing
floors. The program was initiated to complement
the block rotations at JHHA by giving fellows
continuity experience in providing long-term care
over the course of their fellowship.

Out of the need to provide acute and inter-
current care to the residents on the longitudinal
model came the concept of the group-practice fel-
low. The rotation is analogous to a private group
practice, in which physicians jointly care for a
cohort of patients. The group-practice fellow
works closely with a geriatrics nurse practitioner
and attending geriatrician and, in the process,
learns nursing-home medicine as practiced in an
academic nursing home.

Creation of the group-practice rotation al-
lowed another clinical rotation at JHHA to be
dedicated to the rehabilitation unit. This rotation
thus evolved into the most intensive training in
rehabilitation medicine during the fellowship.

During the 1986-1987 academic year, the ro-
tation at the Elmhurst Hospital Center, Queens,
was replaced by a rotation based at the Bronx
Veterans Affairs Medical Center, with support
from the Eastern Paralyzed Veterans Associa-
tion. From its outset, the Bronx VAMC has pro-
vided training and experience in several different
programs and levels of care, including the Nurs-
ing Home Care Unit, the Hospital-Based Home
Care Program, and the unique opportunity to
gain experience with aging spinal-cord-injury pa-
tients by fellows acting as consultants to the Spi-
nal Cord Injury Unit.

1988-1989. National developments in geriat-
rics influenced continued changes in the fellow-
ship program.

The first national Certifying Examination
for Added Qualifications in Geriatric Medicine
was given in April 1988. Also in 1988, the first
round of geriatrics fellowship programs were ac-
credited by the Residency Review Committees for
Internal Medicine and Family Practice of the Ac-
creditation Council for Graduate Medical Educa-
tion. The Special Requirements for Residency
(Fellowship) Training in Geriatric Medicine, de-
lineated in the AMA Graduate Medical Educa-
tion Directory (13) and based on guidelines for
fellowship training programs developed by an Ad
Hoc Committee of the American Geriatrics Soci-
ety (14), helped formalize the components of the
training program.

Within the fellowship program, a new rota-
tion focused on intensive experience in the eval-

Vol. 60 No. 6

FELLOWSHIP TRAINING— PETERSON

471

uation and management of syndromes common
among elderly persons. This rotation utilized the
department's special emphasis clinics and pro-
grams: Osteoporosis and Metabolic Bone Disease,
Falls and Immobility, and Urinary Incontinence.

Another major development in the fellowship
program was the institution of an optional third-
year research fellowship, with support from the
John A. Hartford Foundation.

1990-1991. The 1990-91 academic year
marked major changes in the role of the fellow on
the two major clinical rotations of the second
year.

Geriatrics Evaluation and Treatment Unit. In
response to a mandate from the American Board
of Internal Medicine to incorporate geriatrics
training in internal medicine residencies, a house-
staff team from the Department of Medicine be-
gan rotating on the department's inpatient unit
in July 1990. The fellow on the Geriatrics Eval-
uation and Treatment Unit was then able to as-
sume a junior-attending role as leader of the
team, with an emphasis on performing and teach-
ing inpatient assessment of elderly patients.

Geropsychiatry. Implementation of the recent
regulations in New York State restricting house-
staff hours required an expansion in the size of
Mount Sinai's psychiatry residency program.
This allowed the role of the geriatrics fellow to
change to that of an on-site geriatric medicine
consultant, with an emphasis on assessment and
management of common geriatric syndromes.

Current Status

It can be argued that (at least for geriatri-
cians) the classic academic medicine triad of pa-
tient care, research, and education should be ex-
panded to a tetrad by adding the realm of
administration. This tetrad serves as an outline
for reviewing the current status of the fellowship
program.

Clinical Training

Longitudinal Components. As elaborated
above, the Coffey Geriatrics Clinic at Mount
Sinai and the longitudinal model at JHHA afford
the fellow experience in providing continuity of
care to two different cohorts of older persons.

Block Components. In the standard rotation
schedule (Table 2), the duration for each rotation
is ten to eleven weeks, with the exception of Spe-
cial Emphasis (4—6 weeks) and Clinical Research
n (26 weeks).

At the Bronx VAMC, a 16-bed geriatrics
evaluation and management unit has been cre-

ated, giving fellows experience in assessment of
the elderly during the first year. A recent addi-
tion to the Special Emphasis rotation has been
the home medical care program, in which fellows
acquire experience in the delivery of comprehen-
sive interdisciplinary care to the frail homebound
elderly.

Research

Development of skills in research has been
an objective of the program from its inception. To
accomplish this, fully one-third of the two-year
fellowship is protected research time: a two-and-
a-half-month block rotation in the first year and
six months during the second year.

From August through January, all fellows
attend the Human Aging Research series and the
research methods course, in sequence. During the
first year, the Fellowship Research Committee
meets with fellows individually to explore re-
search interests and to offer assistance to those
needing help in defining projects suitable for im-
plementation. A faculty mentor is also identified
at this time, and fellows are asked to submit a
proposal for their research project. After a review
of the literature, the fellow develops a research
proposal and, if it is approved, begins data collec-
tion.

Six months of protected research time during
the second year provides the fellow with time to
complete the project started during the first year
and in some circumstances to commence work on
another project. Fellows have several opportuni-
ties to present the results of their research pro-
jects (to faculty and colleagues at a department
research seminar; to a larger audience at a Fel-
lows Research Day held in conjunction with their
graduation). At the regional level, fellows partic-
ipate in the annual Fellows Research Night, co-
sponsored by the Metropolitan Area Geriatrics
Society and the Section on Geriatric Medicine of
the New York Academy of Medicine. Many re-
search projects have been accepted for presenta-
tion at the annual meetings of the American Ge-
riatrics Society and The Gerontological Society of
America.

Education

Fellows gain experience as teachers through
teaching medical students and geriatrics nurse
practitioner students during their clinical rota-
tions. In addition, second-year fellows help facil-
itate workshops for first-year medical students
during the geriatrics component of their Human
Development course.

472

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Fellows gain experience in speaking to out-
side professional and lay groups through partici-
pation in the department's Speaker's Bureau. Au-
diovisual resources are maintained for fellows'
use, and they are assisted in learning techniques
for creating new materials.

Administration

Fellows' interest in the administrative as-
pects of health care delivery, including medical
practice and long-term-care administration, con-
tinues to increase. During the past several years,
health policy seminars have been conducted on a
variety of topics. A seminar series addressing ad-
ministrative aspects of geriatric care is currently
being developed.

Outcome of the Program

The "outcome" of any training program is its
graduates. No data comparable to those reported
by Siu et al. (15) are available at this time for
graduates of our program. However, a total of 53
individuals have undertaken fellowship training
over the last decade, and 43 have graduated from
the two-year program. Approximately 60% of the
graduates have remained within the New York
metropolitan area, and 72% hold academic ap-
pointments in a large number of teaching insti-
tutions, both locally and across the country. Ap-
proximately 25% have entered full-time private
practice.

Current and Future Challenges

A debate about the duration and content of
geriatrics fellowship training has recently been
stimulated by a proposal of Dr. William Hazzard
to offer clinical one-year fellowships as an option;
the reader is referred to the original article (11)
and resulting letters (12).

The biggest challenge continues to be the re-
cruitment of high-quality fellows. Comparing
notes with other program directors from across
the country reveals that this problem is not
unique to our department. Several trends contrib-
ute to this phenomenon. First, the applicant pool
is smaller, as fewer medical students are choosing
the two specialties, internal medicine and family
medicine, from which our applicants are drawn.
In addition, residents in internal medicine are
more inclined toward subspecialties that are pro-
cedure-oriented and remunerative. Also, as new
geriatrics fellowship programs have been estab-
lished, the competition for the available appli-
cants has increased.

It is paradoxical that this recruitment situa-

tion prevails, given the abundant professional op-
portunities available to formally trained geriatri-
cians. It is hoped that a national shift in emphasis
toward primary care will result in an improved
outlook for geriatrics fellowship programs and
the field of geriatric medicine. We at Mount Sinai
feel that our experience over the past ten years
clearly demonstrates that well-trained graduates
of a geriatrics fellowship program can be success-
ful in obtaining entry-level positions in major ac-
ademic centers. Through continued tracking of
our graduates, we will be able to determine how
well they succeed in the highly competitive world
of academic medicine and how well they have met
our departmental objective of training the next
generation of leaders in geriatric medicine.

References

1. Libow LS, Cassell CK. Fellowships in geriatrics. Bull NY

Acad Med 1985; 61:547-556.

2. Kane R, Solomon D, Beck J, Keeler E, Kane R. The future

need for geriatric manpower in the United States. N
Engl J Med 1980; 302:1327-1332.

3. Kane R, Solomon D, Beck J, et al. Geriatrics in the United

States: manpower projections and training consider-
ations. Lexington, MA: Lexington Books, D.C. Heath
and Company, 1981.

4. Institute of Medicine Committee on Leadership for Aca-

demic Geriatric Medicine. Report of the Institute of
Medicine: academic geriatrics for the year 2000. J Am
Geriatr Soc 1987; 35:773-791.

5. Department of Health and Human Services. Personnel for

health needs of the elderly through year 2020. Be-
thesda, MD: Pubic Health Service, National Institute
on Aging, Administrative Document, 1987.

6. Reuben DB, Zwanziger J, Bradley TB, et al. How many

physicians will be needed to provide medical care for
older persons? Physician manpower needs for the
twenty-first century. J Am Geriatr Soc 1993; 41:444—
453.

7. Reuben DB, Bradley TB, Zwanziger J, et al. The critical

shortage of geriatrics faculty. J Am Geriatr Soc 1993;
41:560-569.

8. Libow LS. A fellowship in geriatric medicine. J Am Geri-

atr Soc 1972; 20:580-584.

9. Gunby P. Long-term care training, research expanding.

JAMA 1993; 269:2337.

10. American Geriatrics Society. Directory of Fellowship Pro-

grams in Geriatric Medicine, 5th ed. New York: Amer-
ican Geriatrics Society, 1990.

11. Hazzard WR. Geriatric fellowship training: a revisionist

proposal. J Am Geriatr Soc 1992; 40:1175-1177.

12. Geriatric fellowship training (letters). J Am Geriatr Soc

1993; 41:578-583.

13. American Medical Association. Graduate Medical Educa-

tion Directory, 1993-1994. Chicago: American Medical
Association, 1993.

14. Steel K, Applegate W, Barry P, et al. Guidelines for fel-

lowship training programs in geriatric medicine. J Am
Geriatr Soc 1987; 35:792-795.

15. Siu AL, Ke GY, Beck JC. Geriatric medicine in the United

States: the current activities of former trainees. J Am
Geriatr Soc 1989; 37:272-276.

Education, Clinical Services, and Research
On Treating Older People

Clinical Services

The Geriatric Evaluation and
Treatment Unit:

A Model Site for Acute Care of the Frail Elderly, Education,

and Research

Howard Fillit, M.D., and Myron Miller, M.D.

Geriatric inpatient units are not unique to The
Mount Sinai Hospital; rather, they have been an
integral component of geriatric care for over 50
years (1). Throughout the United States, a grow-
ing number of geographically based hospital in-
patient geriatric units are emphasizing different
aspects of geriatric care, including acute care, ge-
riatric assessment, geriatric rehabilitation, and
geriatric psychiatry (2). The Mount Sinai Medical
Center Geriatric Evaluation and Treatment Unit
(GETU) is an acute care geriatric inpatient unit
that represents a structured approach to the com-
plex, interdisciplinary problems of hospitalized
frail elderly patients whose health care needs of-
ten extend beyond the treatment of a single acute
medical illness. Since many of the unit patients
have multiple chronic illnesses, the goal of geri-
atric care is often the restoration and mainte-
nance of function essential to the preservation of
a reasonable quality of life. The GETU also serves
as a model site to foster the development of inno-
vative and enhanced methods of care for the hos-
pitalized frail elderly through clinical research,
and it is a major teaching site for clinical geriat-
rics at The Mount Sinai Medical Center.

The purpose of the unit as a clinical service is
best exemplified by considering two 90-year-old
women. One is essentially healthy, living with

From the Departments of Geriatrics and Adult Development
(HF, MM), Medicine (HF, MM), and Neurobiology (HF),
Mount Sinai School of Medicine and The Mount Sinai Hospi-
tal, New York, NY. Address reprint requests to Dr. Howard
Fillit, Director, Geriatric Evaluation and Treatment Unit, De-
partment of Geriatrics, Box 1070, Mount Sinai Medical Cen-
ter, New York, NY 10029.

her spouse at home. During the flu season, she
becomes superinfected with pneumonia and is ad-
mitted to a traditional hospital bed, is given the
standard 14 days of antibiotic therapy, and goes
home. She does well on a traditional medical
ward.

The other 90-year-old woman lives in a
fourth-floor apartment in a building without ele-
vators. She has become deconditioned, has fallen
in her apartment several times, and has devel-
oped a fear of falling. She has mild cognitive im-
pairment of unknown etiology. She rarely ven-
tures outside and has not been to the doctor in six
months, despite her long history of multiple
chronic illnesses. When this patient becomes ill
with the flu and pneumonia, she also develops
delirium.

On her admission to a hospital, several addi-
tional health care problems are noted that may
affect her outcome, including hypertension, dia-
betes, congestive heart failure, malnutrition, de-
hydration, urinary incontinence of recent onset,
and a stage I pressure sore. As a result of the
agitation secondary to delirium and the history of
falls, she has orders for physical restraints. The
patient, already incontinent of urine, now devel-
ops functional fecal incontinence. In the face of
the restraints, she becomes more agitated, climbs
out of bed, and fractures her hip. The stage I pres-
sure sore soon becomes a stage IV pressure sore,
complicated by osteomyelitis, requiring six weeks
of intravenous antibiotics. Her malnutrition con-
tributes to impaired wound healing and causes a
secondary immunodeficiency, which predisposes
her to sepsis secondary to her osteomyelitis. A
vitamin B-12 deficiency goes undiagnosed, yet

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

475

476

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

may contribute to her progressive cognitive im-
pairment. Even if she does recover from her acute
illness, she may be sent home without maximal
comprehensive nursing, medical, and social ser-
vices, only to return to the emergency room in a
few weeks with more severe malnutrition and a
worsening pressure sore or a fall. Her hospital
length of stay is prolonged by the complications
she suffers as well as the complex discharge plan-
ning that she needs.

The traditional medical ward may not have
the expertise or resources to provide comprehen-
sive and efficient care for such frail elderly pa-
tients with multiple complex interdisciplinary
needs. The GETU is organized to bring interdis-
ciplinary experts and resources together in a geo-
graphically based setting to provide this type of
care to these patients. Indeed, several studies
have demonstrated the benefits of geriatric eval-
uation and management programs for patient
outcomes and health care costs (3).

Criteria for Admission to
The Unit

The unit admits frail elderly people, gener-
ally over age 75, with complex interdisciplinary
health care problems who have the potential to
benefit from its unique resources (Table 1). Ad-
mission criteria have been modified from previ-
ously published criteria for admission to geriatric
assessment programs operating in the Veterans
Affairs but not under Medicare Prospective Pay-
ment (4). The admission criteria to the unit stress
its role as an acute care geriatric unit functioning
under guidelines for Medicare acute care. Several
DRGs, generally underutilized by the traditional
internist, may be employed, such as malnutrition,
urinary incontinence, decubiti, change in mental
status.

The exclusion criteria for the unit further de-
fine its function. Although the unit provides ex-
cellent palliative care to a minority of patients, it
is not a hospice or terminal care unit. Nor is the
unit a "geriatric intensive care unit": patients
needing intensive care, regardless of age, should
go to the intensive care unit. Finally, the unit is
not an alternate-level-of-care unit for patients
who need only social services and do not have
complex multidisciplinary needs.

Unit Operation and Staff

The unit is modeled as a hospital specialty
unit, much like the cardiac care unit or the stroke
unit. The demographics of unit patients (Table 2)

TABLE 1

Admission Criteria, Geriatric Evaluation and Treatment
Unit, Mount Sinai Medical Center

Inclusion criteria: Frail elderly patients requiring
intensive, interdisciplinary comprehensive geriatric
assessment, acute geriatric care, or geriatric specialty
care

• Over the age of 65 years; generally over 75 years old

• meets Medicare Prospective Payment "guidelines" for
acute care; DRGs encompassing "geriatric medicine"
diagnoses, such as "change in mental status" or
"malnutrition," should be considered, particularly for
admissions for geriatric assessment

• has complex, multidisciplinary health care problems:
multiple medical co-morbidities, nursing and social
problems requiring inpatient comprehensive geriatric
assessment

• has specialty problems in geriatric medicine, for example:

• cognitive or affective disorders

• falls, immobility, or gait disorders

• urinary or fecal incontinence

• malnutrition or feeding disorders

• complications of polypharmacy

• pressure sores

• sensory impairments

• sleep disorders

• elder abuse associated with medical co-morbidities

• has potential to benefit from intensive, inpatient
geriatric assessment, rehabilitation, management, or
treatment services to restore function

Exclusion criteria:

• requires hospice or terminal care (patients with severe
cognitive impairment occasionally admitted for acute
geriatric care)

• requires intensive care; should be transferred to the
intensive care unit

• requires only social services (such as nursing home
placement)

encompass the frail elderly who have complex
medical and nursing problems and functional and
psychosocial impairment. Although a majority of
patients are currently admitted via the Emer-

TABLE 2

Summary Demographics: 233 Frail Elderly Patients,
Geriatric Evaluation and Treatment Unit (4187-7188)

Characteristic

Data

Age

80 years

Sex

70% female

Education

8.4 years

Married

25%

Race

45% white

24% black

31% hispanic

Primary insurance

94% Medicare

No. of complicating medical diagnoses

Percent with psychiatric co-morbidity

56%

No. of procedures during hospitalization

Nursing Intensity (MEDICUS scoring)

2.3

Adapted from (7).

Vol. 60 No. 6

EVALUATION & TREATMENT— FILLIT & MILLER

477

gency Room or the Geriatric Clinic, an increasing
number are admitted specifically for geriatric as-
sessment and acute geriatric care from the De-
partment of Geriatrics private practice, as well as
from the Department of Medicine medical wards,
Internal Medicine Associates practice, and pri-
vate practitioners. These referrals are greatly en-
couraged, since the unit is a service for the benefit
of all frail elderly medical patients at The Mount
Sinai Hospital. Although the unit team necessar-
ily functions in the traditional "service" model,
the unit attending physician being the attending
physician during the patient's stay, the input of
each patient's primary care physician is actively
sought and appreciated, and patients discharged
from the unit are always returned to their pri-
mary care physician with a report from the unit
to insure continuity of care after discharge. Only
about 10% of unit cases come from, or are dis-
charged to, nursing homes.

Although the unit is not a high-technology
unit, it does provide intensive, interdisciplinary
acute care to medically complex patients. Medical
care is supervised by attending physicians who
are board certified in internal medicine with
added qualifications in geriatric medicine, and by
geriatrics fellows from the Department of Geriat-
rics and Adult Development. In addition to ad-
dressing the traditional medical problems that re-
sult in the acute hospital admission, the
geriatricians provide geriatric specialty care, in-
cluding, for example, evaluation for reversible
causes of falls or urinary incontinence; investiga-
tion of the etiology of cognitive impairment; man-
agement of decubitus from a medical perspective;
or recognition and treatment of malnutrition. The
geriatrics fellow is responsible for the conduct of
comprehensive geriatric assessment during the
hospital stay. Medical housestaff are responsible
for traditional primary medical care.

Consultants play an important role in the
unit. Since about 50% of our frail elderly patients
have a psychiatric co-morbidity, most commonly
dementia, depression, or delirium, a liaison psy-
chiatrist with board certification in geriatric psy-
chiatry is an integral unit team member. The im-
pact of psychiatric co-morbidities on medical care
is manifest by such events as patient refusal to
cooperate with radiologic tests, drug noncompli-
ance, and falls. The co-morbidities of dementia
and delirium result in an approximate doubling
of hospital length of stay (5, 6). Thus, the presence
of the geriatric psychiatrist not only improves
quality of care, but may insure efficient care and
reduce length of stay. Other specialists, including
a geriatrics neurologist, geriatrics physiatrist.

and a nutritionist, also participate on the unit
team.

The unit is staffed by nurses from the Divi-
sion of Geriatric Nursing with special training in
geriatric nursing. The unit has one of the highest
nursing-to-patient ratios in the hospital because
the average patient carries a high score on the
MEDICUS scoring system employed throughout
The Mount Sinai Hospital to determine nursing
staffing on individual units (6). In fact, the unit
nursing care acuity is almost equal to that in the
cardiac care unit.

Social care is critically important in the unit
(7). Most hospitalized frail elderly have signifi-
cant psychosocial problems that have an impor-
tant effect on health care outcomes. The unit so-
cial worker has special training in geriatric social
work. In contrast to the social worker on the tra-
ditional medical ward, the geriatrics unit social
worker functions following a "screening" rather
than "consultation" model, so every patient is
seen by the social worker within 24 hours of ad-
mission to insure effective social counseling and
efficient discharge planning from day one of ad-
mission. Because family members are often es-
sential to the posthospitalization care of the frail
elderly as caregivers or as case managers, the
unit holds frequent family conferences. House-
staff are asked to participate in these family con-
ferences and are often enlightened to discover the
great efforts families are often required to provide
to meet their frail elderly family member's health
care needs. These conferences also assist in im-
proving quality of care and reducing length of
stay by facilitating discharge planning.

The Process of Geriatric
Assessment in the Unit

Comprehensive and efficient acute geriatric
care is provided through the process of geriatric
assessment combined with traditional medical
care. Comprehensive geriatric assessment is a
method of geriatric care that employs the team
approach to the evaluation and management of
the hospitalized frail elderly with complex inter-
disciplinary health care needs (see box) and is ac-
complished through the use of an organized, effi-
cient set of "instruments" to insure that the
assessment process is structured and quantitative
and does not prolong hospital length of stay.

Geriatric assessment in the unit is conducted
by the unit fellow and is facilitated and organized
by the Handbook for Geriatric Assessment (8).
The admission comprehensive geriatric assess-
ment is completed within 48 hours of admission

478

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Essential Components of Comprehensive
Geriatric Assessment

Physical Health

• geriatric medicine history, physical
exam, laboratory tests; for example, gait/
falls/mobility assessment, incontinence
assessment

Functional Health

• activities of daily living

• instrumental activities of daily living

Mental Health

• cognitive assessment (dementia, delir-
ium)

• affect assessment (depression)

Social Health

• economic assessment

• caregiver assessment

• health care proxy

• estate planning

and summarized by the fellow in a "Geriatric As-
sessment" admission note in the patient's medical
record. The results of the assessment process are
discussed with housestaff, students, and other
team members.

Frail elderly patients may benefit from ad-
mission to the unit for inpatient comprehensive
geriatric assessment when an outpatient assess-
ment would require numerous clinic or office vis-
its over the course of several weeks or months,
during which time the patient might progres-
sively deteriorate as a result of unresolved prob-
lems and be at great risk for further morbidity or
even mortality. Thus, an important concept re-
garding admissions for "geriatric assessment" un-
der Medicare is that of multiple diagnoses in frail
elderly patients for whom the diagnostic sum is
greater than its parts. That is, whereas no single
diagnosis alone might justify the admission of an
individual patient, a multitude of diagnoses
might equal an "acute" admission because the
sum of the complex problems results in a geomet-
ric increase in morbidity. Although obvious risks
are associated with hospitalization, these might
be outweighed by the benefits of a relatively
quick and comprehensive evaluation of complex
problems.

Medicare does not provide a diagnosis-re-
lated group (DRG) for geriatric assessment.
Therefore, the performance of geriatric assess-

ment under the Medicare Prospective Payment
system must be an efficient and comprehensive
process that does not prolong hospital length of
stay. Although these patients clearly require hos-
pital admission to the unit for geriatric assess-
ment, they may not have traditional diagnoses
normally employed by internists for acute care
under Medicare. However, many DRGs of the
Medicare Prospective Payment system can be em-
ployed for admitting these frail patients. In addi-
tion to the usual acute-care diagnoses, a number
of DRG diagnoses are applicable, such as change
in mental status, Alzheimer's disease or senile
dementia, decubiti, urinary incontinence, sleep
disorders, cachexia, malnutrition, or weight loss.
Diagnosis alone is not the sole criterion for ap-
proval of admission. The most important aspect of
a successful Medicare admission for "geriatric as-
sessment" is the documentation in the admitting
note of the medical record, which clearly explains
the acuity of the patient's illness, particularly
from an expert geriatric medicine perspective.

Is the Unit Cost-Effective?

Several research studies have demonstrated
that geriatric assessment units provide high-
quality comprehensive geriatric care while sav-
ing the health care system money (3). With
proper targeting, geriatric assessment units
clearly have the capacity to improve a variety of
hospital outcomes, including reduction in mortal-
ity for patients when followed for up to one year.
Much of the cost savings has come from a reduc-
tion in nursing home utilization during the year
after admission to the geriatric unit.

However, the Mount Sinai unit is a mixed-
use acute-care geriatric unit that functions to
meet the needs of our local hospital health sys-
tem. As with other special care units such as
those for stroke or cardiac care, the actual cost
effectiveness of this geriatrics unit has not been
demonstrated. However, data on length of stay is
available. In 1987, the average length of stay was
approximately 28 days. Subsequent studies (7) re-
vealed the important contribution of hospital "so-
cial stays" to total length of stay in the unit (Ta-
ble 3). These studies demonstrated that a
minority of frail elderly with prolonged hospital
lengths of stay that included up to 70% alternate-
level-of-care days markedly prolonged total
length of stay on the unit. By 1989, the average
length of stay had been reduced to approximately
16 days and now averages about 15 days. Major
reductions in hospital length of stay were
achieved through focused team efforts at dis-

Vol. 60 No. 6

EVALUATION & TREATMENT— FILLIT & MILLER

479

charge planning, including intensive social work
intervention (7). These data are consistent with
other hospital reports of length of stay for frail
elderly patients in this age group (average age 83
years) who commonly have psychiatric co-morbid-
ity and social problems (9). The decrease in length
of stay achieved by the reduction in alternate-
level-of-care stays and other mechanisms was es-
timated to save over $3 million per year in hos-
pital costs. Thus, our hypothesis is that acute
geriatric care in the unit provides a win-win sit-
uation: improvement in quality of care, accompa-
nied by a reduction in hospital length of stay com-
pared to an equivalent "control" group of
hospitalized frail elderly given acute care on tra-
ditional medical wards.

Education in Geriatric Medicine
In the Unit

The unit serves as the primary site for train-
ing in geriatrics for housestaff from the Depart-
ment of Medicine. In this model, housestaff play
an integral role in the provision of primary acute
care while learning the principles of geriatric
medicine through the supervision and assistance
of fellows and faculty of the Department of Geri-
atrics. The education of medical students is also
an important function of the unit. A geriatrics
rotation is required of all medical students at-
tending the Mount Sinai School of Medicine. Ap-
proximately 20% of students receive their clinical
training on the unit. In addition to significant
didactic training, medical students learn to per-
form complete geriatric assessments on aged pa-
tients under the direct supervision of the unit ge-
riatrics fellow. The unit has been demonstrated to
be a successful site for geriatrics education, both
improving clinical skills and increasing knowl-
edge (10).

There are three components to the body of
knowledge in geriatric medicine taught in the
unit: (a) how normal development (aging) affects
the practice of geriatric medicine; (b) the altered
presentation of disease in geriatric patients; and
(c) specific diseases and syndromes particularly
prevalent in old age. Although the effects of aging
on organ function begin, in general, around the
age of 30 or 40, they usually do not become clin-
ically significant until the age of about 75 or 80.
This knowledge becomes critically important in
the provision of acute care to patients over the
age of 75.

Geriatric medicine encompasses diseases
(such as Alzheimer's disease) and syndromes
(such as falls) that occur almost exclusively in old

TABLE .3

Summary Data: Length of Hospital Stay (LOS) for the Frail
Elderly, Geriatric Evaluation and Treatment Unit Days,
Mean ± SD

Pt group (no.)

LOS

Total

Acute ALC

All unit pts (233)

27 ± 31

19 ± 20 —

Acute care only (150)

16 ± 16

16 ± 16 —

Acute care and ALC (83)

48 ± 38

24 ± 25 24 ± 26

ALC, alternate level of care ("social care")
Adapted from (7).

age and are rare in traditional "middle age" med-
icine. These geriatric syndromes often have mul-
tifactorial, sometimes reversible, causes. Falls
provide a good illustration (11). Falling is con-
nected with a high mortality rate in the elderly.
Approximately 20% of people over the age of 80
who fall and break their hip are dead within one
year. In a young population, the most common
cause of falling might be syncope. But the most
common cause of falling in old age is simple de-
conditioning.

Teaching on the unit is provided by geriatrics
fellows and attending physicians on a daily basis
through bedside rounds and through formal talks
and conferences. Specialists in geriatric psychia-
try, neurology, and physiatry also make weekly
specialty teaching rounds. A syllabus with arti-
cles on geriatric medicine from the New England
Journal of Medicine, the Annals of Internal Med-
icine, the Journal of the American Geriatrics So-
ciety, and other geriatric medicine subspecialty
journals, as well as traditional textbooks on geri-
atric medicine (12), expose housestaff to the liter-
ature of geriatric medicine and help prepare them
for the questions on geriatric medicine in the
American Board of Internal Medicine Certifica-
tion Examination.

Clinical Research in the Unit

The unit has been successfully employed as a
site for clinical research in acute geriatric care.
Research on length of stay in the hospitalized
frail elderly has demonstrated the significance of
alternate-level-of-care or "social care" stays in
prolonging total hospital length of stay of these
patients and has demonstrated the critical impor-
tance of intense social work intervention to pre-
vent lengthy social stays (7).

Studies of the impact of dementia on ethical
decisions have demonstrated that the majority of
acutely ill frail elderly medical patients and their
families opted for resuscitation rather than

480

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

forgo life-sustaining technologies, but that for the
demented elderly, families tended to choose "do
not resuscitate" orders more frequently for their
demented family member than patients them-
selves might (13).

Studies of the impact of dementia on acute
medical care in the frail elderly have demon-
strated the interesting finding that, whereas the
most common cause of admission to hospital for
the frail elderly without dementia was, as ex-
pected, cardiovascular disease, the most common
cause of admission to hospital for medically ill
demented patients was infectious disease (6).
These studies also demonstrated the impact of de-
mentia on hospital length of stay and costs (Table
4) and the gross underreporting of dementia as a
DRG diagnosis in medically ill patients (Table 5).
Other studies have compared the utility of two
assessment instruments in the evaluation of cog-
nitive impairment in acutely ill hospitalized frail
elderly patients (14).

Studies of malnutrition and immunologic
function in the hospitalized frail elderly have
shown that over 40% of patients admitted to the
unit were malnourished, that the sensitivity of
detection of malnutrition in these patients could
be increased employing anthropometry, and that
the rate of immunodeficiency in the unit popula-
tion was high, 43% of patients demonstrating
lymphopenia and impaired delayed skin test hy-
persensitivity (15). These data emphasize that
malnutrition is the most common (though often

TABLE 4

Effect of Dementia on Hospital Length of Stay (LOS),
Geriatric Evaluation and Treatment Unit Days, Mean ± SD

Patients Patients
without with
dementia dementia p

Total LOS 17.12 33.55 .001

(2.67) (35.03)

No. of acute days 12.83 23.54 .005

(9.73) (26.82)

No. of alternate-

level-of-care days 4.29 10.01 .029

(8.52) (17.94)

Assigned LOS for DRG 6.59 7.78 .011

003
066

DRG, diagnosis-related group

* Hospital operating costs minus revenues.

Adapted from (6).

TABLE 5

Dementia and the Medical Records Database, Geriatric
Evaluation and Treatment Unit

% of all dementia cases

employed by hospital

for DRG reimbursement

Dementia coding

(N = 90)

Admitting diagnosis

3.3

Co-morbidity

31.1

Total

34.4

Adapted from (6).

unrecognized) cause of acquired immunodeficien-
cy in the elderly (16).

The two oldest known patients with acquired
immunodeficiency syndrome (AIDS) induced by
human immunodeficiency virus (HIV) have been
reported. In one case, a 92-year-old man suffered
an HIV infection at the age of 87 from a transfu-
sion (17). The other case was an 89-year-old
woman with HIV-AIDS secondary to heterosex-
ual transmission (18). These two cases emphasize
the occult nature of HIV-AIDS in elderly patients
with typical geriatric syndromes of dementia and
cachexia.

Summary

Geographically based inpatient geriatric
units are an increasingly common organizational
approach throughout the world for the care of the
hospitalized frail elderly with complex, multidis-
ciplinary acute care needs. The unit provides
comprehensive and efficient quality care to these
patients by bringing together, in a geographically
based unit, specialists from a multiplicity of dis-
ciplines to provide needed services in a coordi-
nated manner. To accomplish its goals, the unit
employs experts in geriatric medicine, nursing,
social work, psychiatry, and other disciplines in a
team approach to comprehensive assessment.
The education of medical students, medical house-
staff, and geriatrics fellows is a critical function of
the unit. The unit also serves as a site for re-
search in improving the care of the hospitalized
frail elderly.

References

1. Matthews DA. Dr. Marjory Warren and the origin of Brit-

ish geriatrics. J Am Geriatr Soc 1984; 32:253-258.

2. Epstein AM, Hall JA, Besdine R, Cumella E Jr, Feldstein

M, McNeil J, Rowe JW. The emergence of geriatric as-
sessment units: the "new technology of geriatrics." Ann
Int Med 1987; 106:299-303.

3. Rubenstein LZ, Stuck AE, Siu AL, Weiland D. Impacts of

geriatric evaluation and management programs on de-

Total LOS

No. of acute days

No. of alternate-
level-of-care days

Assigned LOS for DRG

Difference between
actual LOS and mean
DRG LOS

Net hospital profit
or loss*

17.12 33.55

(2.67) (35.03)

12.83 23.54

(9.73) (26.82)

4.29 10.01

(8.52) (17.94)

6.59 7.78

(2.65) (2.70)

10.53 25.29

(12.33) (34.46)

-$3,331 -$5,910

($7,286) ($9,034)

Vol. 60 No. 6

EVALUATION & TREATMENT— FILLIT & MILLER

481

fined outcomes: overview of the evidence. J Am Geriatr
Soc 1991; 39:8s-16s.

4. Rubenstein LZ, Josephson KR, Weiland GD, English PA,

Sayre JA, Kane RL. Effectiveness of a geriatric evalu-
ation unit: a randomized clinical trial. N Engl J Med
1984; 311:1664-1670.

5. Fulop G, Strain JJ, Vita J, Lyons JS, Hammer JS. Impact

of psychiatric co-morbidity on length of hospital stay for
medical/surgical patients: a preliminary report. Am J
Psych 1987; 144(7 ):878-882.

6. Torian LV, Davidson EJ, Sell L, Fulop G, Fillit H. The

effect of senile dementia on acute medical care in a
geriatric medicine unit. Int Psychogeriatr 1992; 4:231-
239.

7. Fillit HM, Howe J, Fulop G, Sachs C, Siegal P, Sell L,

Miller M, Butler RN. Studies of hospital social stays in
the frail elderly and their relationship to the intensity
of social work intervention. Soc Work Health Care
1992; 18:1-22.

8. Fillit H. Handbook for geriatric assessment. Available on

request.

9. Safran DG, Eastwood EA. Transitional care: the problem

of alternate level of care in New York City, New York:
United Hospital Fund of New York, 1990.
10. Fields SD, Jutagir R, Adelman RD, Tideiksaar R, Olson E.
Geriatric education. Part I: Efficacy of a mandatory

clinical rotation for fourth year medical students. J Am
Geriatr Soc 1992; 40:964-969.

11. Tideiksaar R. Falls and gait disorders. In: Abrams WB,

Berkow R, eds. The Merck manual of geriatrics. West
Point, PA: Merck Sharp and Dohme Research Labora-
tories, 1990:52-68.

12. Brocklehurst JC, Tallis RC, Fillit HM. Textbook of geri-

atric medicine and gerontology, 4th ed. London: Chur-
chill Livingstone, 1992.

13. Torian LV, Davidson EJ, Fillit HM. Decisions for and

against resuscitation in an acute geriatric medicine
unit servicing the frail elderly. Arch Int Med 1992;152:
561-565.

14. Fields SD, Fulop G, Sacks C, Strain J, Fillit HM. A com-

parison of two instruments for screening for cognitive
impairment. Int Psychogeriatr 1992; 4:93-102.

15. Lansey S, Waslien C, Mulvihill M, Fillit H. The role of

anthropometric assessment for malnutrition in the hos-
pitalized frail elderly. Gerontology (in press).

16. Fillit HM. Reversible acquired immune deficiency in the

elderly: a review. Age 1991; 14:83-89.

17. Fillit H, Fruchtman S, Sell L, Rosen N. AIDS in the el-

derly. Geriatrics 1989; 44:65-70.

18. Rosenzweig R, Fillit H. Probable heterosexual transmis-

sion of AIDS in an aged woman. J Am Geriatr Soc 1992;
40:1261-1264.

The Phyllis and Lee Coffey Ambulatory

Care Clinic:

Ten Years of Evolution in Treating Older Adults

Barbara Paris, M.D., Diane E. Meier, M.D., Jane Morris, R.N., Nurit Ginsberg, C.S.W., and

Linda Weiss, M.P.A.

A TREMENDOUS NEED exists for Outpatient facili-
ties to address the complex medical and psycho-
social problems faced by the frail elderly (1). Such
patients are best served by an interdisciplinary
team that addresses the multidimensional prob-
lems of aging and aims to preserve the functional
independence of the patients in their community.
Our goal was to establish a clinic that could pro-
vide such services by staffing it with an interdis-
ciplinary team of professionals trained in geriat-
rics to provide a longitudinal model for patient
care.

The team approach, a central principle of ge-
riatrics, is a departure from the traditional doc-
tor-centered medical model and is the single most
important characteristic differentiating the geri-
atrics clinic from a standard medical outpatient
clinic. Such an approach is necessary to meet the
needs of the frail older adult patient, whose re-
quirements frequently exceed the skills and re-
sources of any single professional discipline.

After a review of the literature describing pa-
rameters for geriatric clinics, we set in place a
clinic that we hope serves as a model for other
institutions. The clinic is also a key component of
the Department of Geriatrics and Adult Develop-
ment programs associated with housestaff train-
ing, medical student education, and clinical re-
search.

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, New York City. Address
reprint requests to Barbara Paris, M.D., Department of Geri-
atrics, One Gustave L. Levy Place, Box 1070, New York, NY
10029.

Demographics

Established in 1983, the Phyllis and Lee Cof-
fey Ambulatory Care Clinic provides comprehen-
sive care and consultative services to patients
who are primarily 65 years of age or older, many
of them frail. Over 42% come from the surround-
ing East Harlem neighborhood. The population
served ranges in age from 64 to 107 years; the
average age of clinic patients is 79 years. Eighty-
three percent of patients are women; 41% are Af-
rican-American, 26% are Latino, and 12% are
white. Approximately 40% of our patients are
cognitively impaired and 60% are functionally de-
pendent. Many are living alone.

There are 10 to 12 new patient referrals per
week from diverse sources, including the emer-
gency room, hospital inpatient services, commu-
nity agencies, private practices, and word of
mouth (Fig,). Approximately 25 patients are seen
per clinic half-day (for a total of about 80 patients
per week in four clinic sessions, or 3,200 patient
visits per year). The complex and multiple needs
of the patients makes this low number of visits
necessary; a visit may last as long as two to three
hours, requiring the time and attention of several
members of the team and thus prohibiting larger
patient loads. It is hoped that these time- and
effort-intensive preventive measures will save re-
sources over the long term by reducing both the
number and duration of hospitalizations and
nursing home placements.

The most common diagnoses in the Coffey
Clinic mirror those nationally for patients 65
years and older, according to the 1992 National
Center for Health Statistics: hypertension,
chronic heart disease, diabetes mellitus, and os-

482

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

AMBULATORY CARE CLINIC— PARIS ET AL.

483

teoarthritis (2). These have not changed since
Calvert and Koch's report of 1981 (3).

Ninety-seven percent of our patients are cov-
ered by Medicare; the 20% co-payment is charged
on a sliding scale according to ability to pay, and
57% also have Medicaid insurance. The remain-
der pay for their visits according to a sliding fee
scale determined by income and family size. No
patient is turned away for lack of resources.

OUTSIDE REFERRALS

OTHER
CLINICS

EMERGENCY
ROOM

1 /

GERIATRIC
OUTPATIENT
CLINIC

GERIATRIC MEDICAL
HOME CARE PROGRAM

INPATIENT GERIATRIC
CONSULTATION AND LIAISON
SERVICE

GERIATRIC EVALUATION
AND TREATMENT UNIT

The Clinic Team

The clinic team has evolved into a core staff
of over 20 health care providers. The medical staff
on the clinic team includes a geriatrician chief,
attendings in geriatric medicine, and 10 fellows
in a two-year training program. Additional med-
ical consultants in neurology, neuropsychology,
gynecology, gait disturbances, urinary inconti-
nence, and cardiology provide the clinic with
their expertise once a week. The nursing staff in-
cludes three nurses with advanced education in
geriatrics and a nursing assistant, supervised by
the director of geriatric and rehabilitation nurs-
ing. There are three certified social workers and a
supervisory social worker. Other support staff
members include a patient care coordinator, a
Spanish translator, a database coordinator, and
ambulatory care registrars and supervisor.

Patients are assigned to a miniteam consist-
ing of a fellow, a nurse, a social worker, and a
data input person. The team meets frequently to
ensure communication about each patient and to
outline and implement care plans. The entire
clinic staff meets biweekly to discuss the most
complex cases. Ongoing meetings between team
members, patients, and caregivers assure clarity
of communication and permit all involved to work
together toward shared goals. Two additional reg-
ularly scheduled bimonthly meetings facilitate
this process: a team meeting of all clinic staff, and
an operational group meeting of the clinic chief,
the director of geriatric nursing, and the social
work supervisor. The more specific roles of team
members are as follows.

Attending physicians are responsible for ad-
ministrative issues that arise during clinic ses-
sions, for medical consultation with the fellows,
nurses, and social workers, and for teaching med-
ical students.

Fellows serve as primary care and consulta-
tion physicians for all clinic patients. Each fellow
is assigned a panel of approximately 50 patients
and follows those patients throughout fellowship
training.

OTHER INPATIENT
UNITS

Fig. Sources of patient referrals.

The clinic chief, the director of geriatric
nursing, and the social work supervisor are re-
sponsible for general administration and policy of
the clinic and for support and development of the
team.

Geriatric nurses assess the patient's and the
family's educational needs and teach preventive
health measures (such as medications, nutrition,
safe transferring from bed to wheelchair); they
also provide highly skilled assessment and treat-
ment of some of the most difficult problems of
clinic patients, including pressure ulcers and
bowel and bladder incontinence. The nurses also
serve as liaisons to multiple community agencies.

Nursing assistants work under the immedi-
ate direction of the geriatric nurses and assist all
members of the health care team. The nursing
assistants' observations of patients are shared
with other team members.

Social workers perform an initial psychoso-
cial assessment of the patient and the family,
evaluating coping capacities and available sup-
port networks and monitoring for depression, iso-
lation, and elder abuse. Their role may involve
crisis intervention and ongoing individual, fam-
ily, and group therapy to help sort out conflicts
arising from the needs of the older patient, in-
cluding bereavement, anxiety over nursing home
placement, elder abuse, and problems associated
with Alzheimer's disease. Social workers also as-
sist patients in identifying and acquiring commu-
nity health and social services. The physician-
nurse-social worker triad schedules frequent
meetings to ensure timely communications about
changes in the patient's status and appropriate
adjustments of the care plan.

The neuropsychologist serves as an addi-
tional resource for patients' psychological and
neuropsychological problems, available to discuss
mental status abnormalities and questions about
psychiatric diagnoses, psychotherapy, and family

484

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Initial Outpatient Geriatric Assessments

Nursing Assessment

medical history

medications

nutrition

skin

functional status
mental status
mobility

health maintenance
protocols

Medical Assessment
physical examination
advance directives
dementia
depression
osteoporosis
peripheral arterial

obstructive

disease
sleep disorders
urinary incontinence
substance abuse
elder abuse and

neglect
sexual functions

Social Work Assessment

financial

psychological

family history

mood

caregiver burden

management with the fellow-nurse-social worker
team.

The patient care coordinator screens and
schedules all new patient appointments, assists
with transportation arrangements for patients,
makes reminder phone calls to patients prior to
their appointments, and follows up on all patients
who miss their appointments. An important role
is monitoring the status of patients to identify
those who are hospitalized, transferred to a
chronic care facility, lost to follow-up, or dead.

The Spanish translator interprets for pa-
tients whose primary language is Spanish (30% of
the clinic population). She also functions as a pa-
tient advocate. The staff relies on her expertise to
identify the complexities of bilingual and bicul-
tural communication. Patients often communi-
cate information critical to their health status to
the translator in the course of casual conversation
in the clinic waiting area.

The database coordinator enhances commu-
nication between the department's various clini-
cal services by assuring that up-to-date, vital in-
formation on medications, diagnoses, and health
maintenance is available on the patient's medical
record. Information is updated after each clinic
visit.

The ambulatory care registrars, under the
guidance of their supervisor, are receptionists in
the clinic whose functions include making follow-
up appointments for patients and medical record
keeping.

The Team Approach to
Patient Care

Geriatric assessment is the usual initial ap-
proach to gathering data on each patient and for-
mulating a care plan. The goals of the assessment
are

• Improved diagnostic accuracy

• Improved functional status or rehabilitation

• Improved efficacy of and compliance with med-
ications

• Health promotion and disease prevention

• Better use of community support services

• Improved emotional status and sense of well-
being

• Reduced use of acute and long-term care facil-
ities

The average evaluation requires 3-5 visits to the
geriatric clinic and results in a detailed and thor-
ough medical, nursing, and social assessment, di-
vided among the various members of the team
(see box).

Each new patient is seen in an initial screen-
ing and assessment clinic staffed by the nurses,
social workers, Spanish translator, and patient
care coordinator. Validated assessment tools are
used to evaluate functional status (4, 5), mental
status (6, 7), mood (8), mobility (9), and caregiver
stress (10). Additional data is gathered on medi-
cal history, skin care, nutrition, medications, fi-
nances, psychosocial circumstances, abuse and
neglect.

Health maintenance protocols, based on the
Report of the United States Preventive Services
Task Force (11), are initiated, including blood and
urine tests, electrocardiograms, tuberculosis
screening, dental, vision, and hearing screens,
mammograms, fecal occult blood testing, and im-
munizations. Laboratory and radiographic stud-
ies are ordered judiciously, with attention to ex-
pected yield and safety, but no patient is denied a
thorough evaluation because of chronological age.

A patient is given a follow-up appointment
with a physician within two weeks of the initial
clinic visit. The physician then reviews all of the
data collected in Management Clinic and per-
forms additional assessments as individually in-
dicated.

After the initial comprehensive assessment.

Vol. 60 No. 6

AMBULATORY CARE CLINIC— PARIS ET AL.

485

patients may choose to return to their community
physician for primary care or remain in the cHnic.
The vast majority continue to use the clinic as
their primary health care provider.

Patients in the clinic are referred to other
specialists as deemed necessary by the team. In
addition to the many specialists on the clinic
staff, patients have access to rehabilitation med-
icine, speech pathology, and numerous other med-
ical and surgical sub-specialty clinics in close
proximity to the Ambulatory Care Clinic. This
"one-stop" approach to the multiple interacting
problems of the aged contributes to patient com-
pliance and involvement with their care and thus
greatly increases the efficacy of care. This is ad-
vantageous to the patient, for it enables the team
to oversee the patient care received from a variety
of sources. For example, adverse drug reactions
related to polypharmacy are more easily identi-
fied and averted. The team thus becomes the pa-
tient's advocate as he or she enters the larger
health care arena.

Overall, 17% of clinic patients are admitted
to the hospital annually. Most of these are admit-
ted to the Geriatric Evaluation and Treatment
Unit, staffed by attending physicians and fellows
in the Department of Geriatrics and Adult Devel-
opment. Both the inpatient and outpatient clinic
staff meet monthly to discuss patients who have
complex problems and prolonged or repeated hos-
pital admissions. Clinic patients requiring hospi-
talization on other specialty units are followed up
by the Geriatric Consult Service, also staffed by
department faculty and fellows. Outpatients who
become acutely ill when the clinic is not in session
are referred to the hospital emergency room, and
the Geriatric Consult Service is notified to make
an assessment. Clinic patients who become home-
bound are referred to the Geriatric Medical Home
Care Team for continuous care at home.

The Team at Work: A Case Study. An 81-
year-old widow was referred to the clinic by a so-
cial worker at a local senior center for general
medical care. The patient had been widowed for
many years, her only son had died at the age of 30
years, and she had no other relatives. She was
living alone and had little contact with the out-
side world other than one "friend" at the senior
center where she ate lunch every day.

The patient arrived at the clinic escorted by a
volunteer from the senior center. She was un-
kempt, with notably poor hygiene. At her first
clinic visit, she was assessed by a fellow, a nurse,
and a social worker. On initial evaluation, prob-
lems identified included malnutrition, unsteady

gait, and confusion. For example, she could not do
simple arithmetic, and she was unsure of the
date. It was clear that she was incapable of han-
dling her own finances, which she stated her
"friend" took care of for her.

The team discussed the case, and the medical
and psychosocial problems were immediately ad-
dressed. The patient was referred to a community
agency that could provide home attendant ser-
vices, with both nursing and social work supervi-
sion. Ambulette arrangements were initiated for
her future clinic visits. Subsequently, a nurse and
a social worker doing a home visit identified a
need for heavy-duty cleaning services, evidence of
financial mismanagement, limited food, and dirty
clothing. A home attendant was placed in the
home for six hours a day to assist the patient with
cleaning, shopping, bathing, and meal prepara-
tion. The community agency also took the respon-
sibility for future management of the patient's
finances.

The patient continued to come to the clinic,
where her medical and cognitive problems were
further evaluated and treated. She remained sta-
ble in the community setting for eight years. At
age 89, she required hospitalization and surgery
for an acute bowel obstruction. After prolonged
hospitalization, she was too frail and weak to re-
turn home and was transferred to an inpatient
rehabilitation program. Finally, after many
months of reconditioning, the patient's physical
condition improved, and she was very anxious to
go home. With the dedicated efforts of the team,
in coordination with the community agency, this
patient was discharged from the nursing home
and reinstated in her own apartment. At age 90,
she is happy and doing well.

Education and Training

The clinic is an important training site for
fourth-year medical students and fellows. The ed-
ucational objectives include:

Providing exposure to primary and consulta-
tive outpatient care of the older adult in a setting
emphasizing continuity of care.

Learning the basic principles of case manage-
ment and coordination of multiple services and
providers in an outpatient setting.

Being comfortable and facile with the team
approach to geriatric care by becoming aware of
the skills and knowledge specific to the disci-
plines of nursing, social work, and psychology so
that appropriate triage, referral, and communica-
tion will take place.

486

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Becoming cognizant of the difficulties and
frustrations that may confront the health profes-
sional in caring for the elderly and learning how
to handle these feelings in a manner that does not
interfere with appropriate high-quality patient
care.

Learning the database of ambulatory geriat-
ric medicine, including the approach to health
maintenance, atypical presentation of disease,
diseases characteristic of the elderly, problems of
therapeutics and polypharmacy, risk-benefit
analyses of diagnostic and therapeutic ap-
proaches to the older adult, and history and phys-
ical examination of the older person.

Being aware of the necessity for comprehen-
sive functional assessment of the elderly patient
for appropriate service assignment, prognostica-
tion, and protection of patient independence.

Becoming accustomed to working with pa-
tients' families and friends to maximize the well-
being, autonomy, and strength of the patient.

Becoming familiar with the language and
concepts relevant to the ethical dilemmas inher-
ent in the care of the elderly, including problems
of autonomy and paternalism, informed consent,
and decisions to forgo life-sustaining therapy;
specific training in how to approach discussions
with patients about advance directives for health
care.

Acting as teachers and role models for stu-
dents from various disciplines rotating through
the clinic, both in direct patient care and in lead-
ership of educational and patient-family confer-
ences.

Becoming familiar with the health-policy is-
sues complicating the care of the elderly by ob-
serving first-hand the difficulties encountered by
clinic patients and their families.

Learning the fundamentals of outpatient
neuropsychological assessment and how to gener-
ate an appropriate differential diagnosis, diag-
nostic plan, and triage-referral pattern for the
cognitive problems of the elderly.

Clinic Research

The clinic also serves as a research setting for
investigators from the Department of Geriatrics
as well as from the larger Medical Center com-
munity. Projects have included a time and inten-
sity study for the Physician Payment Review
Commission, a study of service needs of East
Harlem Hispanic elderly, and the efficacy of ini-
tiation of health maintenance protocols by nurses
over physicians. Results of these research projects
have been presented at national meetings. Cur-

rently, an ongoing two-year study is assessing the
utility of the New York State Health Care Proxy
Law for our clinic population, many of whom have
few social supports.

Ongoing Team
Development Process

Over the past year, with the assistance of
consultants from The Mount Sinai Hospital De-
partment of Human Resources, our interdiscipli-
nary team underwent an intensive review of its
strengths and weaknesses, with the goals of max-
imizing good patient care, identifying and resolv-
ing interprofessional conflicts, enhancing com-
munication skills, and increasing team member
satisfaction in the work place. Three major areas
of concern — the patient appointment system, the
medical records and radiology retrieval systems,
and the team meeting structure — were identified
as requiring the establishment of problem-solv-
ing interdisciplinary working groups for further
analysis. A significant outcome of this process has
identified the need for an overall clinic adminis-
trator whose role would be to oversee clinic oper-
ations and keep the team focused on their com-
mon goal of quality patient care.

Conclusions

In summary, the Coffey Geriatrics Ambula-
tory Care Clinic is organized on an interdisciplin-
ary model to provide comprehensive longitudinal
care to the community's frail elderly, allowing
them to maintain a maximal degree of functional
independence. It also serves as a training site for
housestaff and medical students to learn the
skills needed to care for the frail elderly. To un-
derstand the growing challenges and complexi-
ties of providing such care, the clinic also serves
as a forum for clinical research.

Acknowledgments

The clinic was developed with support provided by Phyllis and
Lee Coffey, Dr. Raymond and Hannah Schneider, and the
Helen and Max Abrams Fund.

References

1. Yeo G, Ingram L, Shurnick J, Crapp L. Effects of a geri-

atric clinic on functional health and well-being of el-
ders. J Gerontol 1987; 42:252-258.

2. U.S. Dept. of Health and Human Services. Vital and

health statistics: health data on older Americans.
Washington, DC: U.S. Government Printing Office,
1992; 125.

3. Calvert JC, Koch HK. Ambulatory geriatric care: a na-

tional perspective. In: Steel K, editor. Geriatric educa-
tion. Lexington, Mass.: Collamore Press, 1981:103-109.

4. Lawton MP, Brody EM. Assessment of older people: self-

Vol. 60 No. 6

AMBULATORY CARE CLINIC— PARIS ET AL.

487

maintaining and instrumental activity of daily living.
Gerontologist 1969; 9:179-186.

5. Katz S, Downs TD, Cash HR, et al. Progress in develop-

ment of the Index of ADL. Gerontologist 1970; 10:20-
30.

6. Folstein MF, Folstein S, McHugh PR. Mini-mental state:

practical method for grading cognitive state of patients
for the clinician. J Psychiatr Res 1975; 12:189-198.

7. Hachinski VC, Iliff LD, Zilhka E, et al. Cerebral blood

flow in dementia. Arch Neurol 1975; 32:632-637.

8. Yesavage JA, Brink TL, Rose TL, et al. Development and

validation of a geriatric depression rating scale: a pre-
liminary report. J Psychiatr Res 1983; 17:37-^9.
9. Tinetti ME. Performance-oriented assessment of mobility
problems in elderly patients. J Am Geriatr Soc 1986;
34:119-126.

10. Zarit SH, Reever KE, Bach-Peterson J. Relatives of the

impaired elderly: correlates of feelings of burden. Ger-
ontologist 1980; 20:649-655.

11. Report of the United States Preventive Services Task

Force. Guide to clinical preventive services: an assess-
ment of the effectiveness of 169 interventions. Fisher
M, ed. Baltimore: Williams & Wilkins, 1989.

Health Care for the Homebound

Older Adult:

A Medical Model

Laura Di Pollina, M.D., Gabriel Gold, M.D., and Diane E. Meier, M.D.

More than 5 million elderly people living at
home are functionally restricted because of mo-
bility limitations; 1.6 million of these are severely
impaired, and 300,000 live alone (1). Although
physician house calls were commonplace before
World War II, they have declined markedly over
the past 50 years, representing only 1.7% of phy-
sician-patient contacts in 1979 (2). Because the
homebound elderly are often in near-total isola-
tion from health-care services, they generally
have no physician and do not contact the health-
care system until they are brought to the emer-
gency room with a severe acute illness. After they
are discharged from a hospital, they are at high
risk of readmission because they are unable to
make use of medical services outside their homes;
not uncommonly, they are discharged to nursing
homes.

In February 1991, the Department of Geriat-
rics and Adult Development at The Mount Sinai
Medical Center initiated a geriatric medical
home-care program (GMHCP) to provide compre-
hensive health-care team management, including
physician visits to homebound elderly. The spe-
cific goals of the home-care program are to reduce
emergency room visits and hospitalizations, obvi-
ate the need for nursing home placement, allevi-
ate burdens on caregivers, provide home care to
terminally ill patients, allow terminal patients to
die at home rather than in the hospital if they so
wish, generally improve the health and quality of

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, New York, New York. Ad-
dress reprint requests to Laura Di Pollina, M.D., The Mount
Sinai Medical Center, One Gustave L. Levy Place, Geriatrics/
Box 1070, New York, NY 10029.

life of the homebound elderly, and provide a
teaching model for fellows, residents, and stu-
dents.

Description of the Program

The program is staffed by a full-time nurse
practitioner, a part-time (30%) social worker, and
a part-time (60%) geriatric physician. Although
the physician is fluent in both English and Span-
ish, the nurse and the social worker speak only
English, and an individual available for Spanish
translation when necessary has been identified.
The program is also integrated with the Mount
Sinai Home Care Agency, which provides other
patient services, such as a nurse coordinator,
physical therapists, occupational therapists, and
escorts to unsafe areas.

Service components include medical care,
nursing, social work, physical therapy, speech
therapy, occupational therapy, podiatry, home
health aides, laboratory services. X-rays, elec-
trocardiograms, Holter monitoring, medical
equipment, and supplies. A physician and a nurse
coordinator are available by telephone 24 hours a
day, seven days a week, to handle emergencies.
Services are planned and coordinated by the pro-
gram and the Mount Sinai Home Care Agency
and are provided either directly or through con-
tractual arrangements.

The initial evaluation is conducted by a
board-certified geriatrician and a registered
nurse and lasts approximately two hours. The ini-
tial social work evaluation is conducted at a later
date (see box). Patients are seen by the physician
every one to two months, or more frequently if
necessary; nursing visits generally occur every

488

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

HOMEBOUND HEALTH CARE— DI POLLINA ET AL.

489

Initial Patient Evaluation for Geriatric
Medical Home-Care Program, Mount
Sinai Medical Center

Medical and Nursing Evaluation

Complete history and physical examination
Housing and environmental assessment
Medication review (prescription and over-
the-counter)
Performance Oriented Environmental Mo-
bility screen (3)
Need for durable medical equipment (hospi-
tal bed, wheelchair, assistive devices)
Functional assessment, adapted from OARS
(4)

Nutritional assessment

Folstein Mini-Mental State Exam (5)

Yesavage Mood Assessment Scale (6)

Social Work Evaluation

Psychosocial screen

Evaluation of caregiver stress by Burden In-
terview Scale (7)

Advance directives (living will, health care
proxy)

Burial plan

Financial status and Medicaid eligibility

one to two weeks, depending on the patient's
needs (for instance, prefilling of medication dis-
penser, blood glucose monitoring, dressings, in-
jections).

Patient Demographics

Since its inception in February 1991, the pro-
gram has received 98 referrals and enrolled 46
patients who met admission criteria, as follows:

• Homebound as a result of physical frailty, cog-
nitive dysfunction, or environmental barriers

• Lacks access to health care

• Resides within catchment area

• Willing to participate, has a support person
available, and is able to summon help in case of
emergency

Thirty patients are currently active. The patient
population is mostly female and of advanced age
(Table), Although whites predominate (52%),
there is a large Hispanic group (30%), all of whom
speak only Spanish; 17% are black. Most patients
were referred from an outpatient setting (53%)
because they were no longer able to keep their
clinic or office appointments as a result of increas-
ing disability; 20% were referred by community
agencies as individuals isolated at home with no
access to a physician's care; 18% were self-re-
ferred; and 9% were hospital inpatients referred
through the Geriatrics Evaluation and Treat-
ment Unit or the Geriatric Consultation and Li-
aison Service. All medical home-care program pa-
tients are Medicare beneficiaries, and 61% are
enrolled in Medicaid.

Social Supports. Of the 46 patients enrolled
in the program, 36 are widowed, 7 are married, 2
are single, and 1 is divorced. Most (48%) live
alone, 15% live with their spouse, and 37% live
with another family member, almost always a
daughter. Many benefit from home health aide
services (Fig.), and 54% of those living alone re-
ceive 24-hour home health aide support.

Functional Status. Patients are generally
dependent in many activities of daily living
(ADD, including feeding, dressing, grooming,
transferring, bathing, and toileting; the average
ADL score for this patient population is 5 (worst
0, best 12). Patients are generally not able to
carry out any instrumental activities of daily liv-
ing (lADL), such as traveling, shopping, meal
preparation, housework, handling of finances, or
self-administration of medications; the average
lADL score of the patient population is 1 (worst 0,
best 12). More than one-quarter of the patients

(28%) are incapable of leaving their home, even
with ambulation assistive devices or the assis-
tance of another person.

Health Status. Enrolled persons are frail and
elderly and have many diseases and disabilities.
The average number of medical diagnoses per pa-
tient is 7 and ranges from 2 to 14; the average
number of medications per patient is 6 and ranges
from 1 to 12. The prevalence of dementia is high
(72%), and the average Folstein Mini-Mental
State score for all patients is 15 (best score 30).
The most common diagnoses include dementia,
osteoarthritis, hypertension, coronary artery dis-
ease, congestive heart failure, hearing impair-
ment, vision impairment, diabetes mellitus, ane-
mia, cerebrovascular disease, and urinary
incontinence. Of the 17 patients discharged thus
far from the program, 50% died at home, 19% died
in hospital, 19% were transferred to nursing
homes, 6% were lost to follow-up after a hospital-
ization, and 6% recovered enough independence
to return to ambulatory clinical services. A re-
markable 73% of all deaths occurred at home
(only 27% in the hospital), thus accomplishing
one of the goals of the program, which is to pro-

490

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

TABLE

Age and Sex Distribution of Patients in the Health Care
Program for the Homebound Elderly

No. of

Mean age

Age range

pts (%)

(SD)

(years)

Women

38 (83)

86 (10)

63-104

Men

8 (17)

83 (8)

66-91

Total

46(100)

85 (9)

63-104

vide supportive and medical services to termi-
nally ill patients who prefer to die at home rather
than in the hospital.

Education of Physicians in
Home Health Care

Medical students and residents generally re-
ceive no training in home care and, as a rule,
their physician role models and mentors do not
visit patients in their own home. The American
College of Physicians (8), the American Geriat-
rics Society (9), and the American Medical Asso-
ciation's Council on Scientific Affairs (10) have
all underscored the need for educating physicians
in home care. Unfortunately, the lack of appro-
priate training sites and mentors is a major hin-
drance to this educational effort. The program is
unusual in that the home-care team includes a
faculty attending physician, thus providing an
appropriate setting for the education of medical
trainees in home care. All 10 geriatric fellows in
Mount Sinai's program spend one month on a spe-
cial rotation that includes home visits with the
program, and many medical students have par-
ticipated in the program as well. This teaching
role is expanding, and by 1994 the program will
be providing training to 30 medical students each
year.

PERCENT OF PATIENTS

30%

10%

NONE 4 HOURS 8 HOURS 12 HOURS 24 HOURS

DAILY HOME HEALTH AIDE PRESENCE

Fig. Proportion of patients of the Geriatric Medical Home-
Care Program at Mount Sinai Medical Center receiving home
health-aide services.

Illustrative Cases

Case 1. A 69-year-old black man was referred
to the program by a medical primary-care clinic.
He could only travel by stretcher and had not
been seen in the referring clinic for over a year.
Acute medical illnesses were requiring almost
monthly emergency room visits.

The initial program evaluation revealed a
patient who was bed-bound as a result of multiple
strokes and who suffered from severe multi-in-
farct dementia, hypertension, pseudobulbar palsy
with dysarthria and dysphagia resulting in recur-
rent aspiration pneumonias, and peptic ulcer dis-
ease that had led to multiple episodes of gastro-
intestinal bleeding and iron-deficiency anemia.
The patient was fed through a gastrostomy tube,
and his mobility was further impaired by a left
lower extremity amputation above the knee. The
patient's wife was overwhelmed by the task of
caring for her husband and at a loss about where
she could turn for help. The home-care program
initiated a rehabilitation program that enabled
the patient to regain enough strength and range
of movement to be able to sit in a wheelchair. The
program social worker addressed his wife's con-
cerns. The patient has been under the care of the
program for almost two years and, in that time,
has visited the emergency room only once, to re-
place a dislodged gastrostomy tube. Occasional
episodes of aspiration pneumonia have been suc-
cessfully treated at home.

Case 2. A 90-year-old woman with multi-in-
farct dementia, chronic atrial fibrillation, hyper-
tension, congestive heart failure, diabetes melli-
tus, a seizure disorder, and a history of multiple
strokes had been receiving care at the geriatrics
outpatient clinic but could no longer be trans-
ported there other than by ambulance. Control of
her diabetes and hypertension deteriorated, and
she was repeatedly brought to the emergency
room for seizures. She and her family were ada-
mantly opposed to nursing home placement. The
case was referred to the home-care program.

The program reviewed diet and compliance
with prescribed medication regimens with the pa-
tient's caretaker (her daughter), and after evalu-
ation of the home situation, it was decided that
the patient could remain home with appropriate
support, including a 24-hour home health aide.
The patient has been able to remain home, receiv-
ing continued medical care through the program.
Weekly nurse visits were initiated to prefill the
patient's medication dispenser, monitor her blood
pressure and diabetic control, and counsel care-
givers. Her phenytoin level has remained within
the therapeutic range, and she has suffered no

Vol. 60 No. 6

HOMEBOUND HEALTH CARE— DI POLLINA ET AL.

491

further seizures; her diabetes and hypertension
are well controlled. The patient's course has been
marked by several episodes of aspiration pneumo-
nia, which were treated successfully at home by
the program, and by another stroke, which led to
a brief hospitalization. She has returned to her
apartment and is receiving oral anticoagulant
therapy with home monitoring of her prothrom-
bin time.

Summary

The medical home-care program is a logical
and indispensable addition to the comprehensive,
multilevel geriatric care offered by the Depart-
ment of Geriatrics. The home-care program en-
ables the frail and homebound elderly to remain
in their home environment and continue to re-
ceive comprehensive primary medical care and
supportive services. It also serves as a valuable
teaching site for medical trainees. Program eval-
uation through assessment of patient and care-
giver satisfaction, estimates of costs of care, and
frequency of hospitalization and emergency room
visits are ongoing.

Acknowledgments

The authors acknowledge the help of Linda Weiss, M.P.A.,
Merlene Godfrey, R.N., Jane Morris, R.N., Shirley Kudalsky,
R.N., and Lynn Harmet, M.S.W.

The geriatric medical home-care program at The Mount
Sinai Medical Center is supported, in part, by a grant from the
New York State Primary Care Initiative.

References

1. Commonwealth Fund Commission on Elderly People Liv-

ing Alone. Help at home: long-term care assistance for
impaired elderly people. Baltimore: The Commission,
1989.

2. Keenan JM, Hepburn KW. The role of physicians in home

health care. Clin Geriatr Med 1991; 7:665-675.

3. Klein W, Taylor B, Tsai T, Tideiksaar R. Reliability of a

performance-oriented environmental mobility screen
for hospitalized elderly. J Am Geriatr Soc 1992;
40:SA27.

4. Fillenbaum GG. Multidimensional functional assessment

of older adults: the Duke Older Americans Resources
and Services procedures. Hillsdale, NJ: Lawrence Erl-
baum Associates, 1988.

5. Folstein MF, Folstein S, McHugh PR. Mini-Mental State:

practical method for grading cognitive state of patients
for the clinician. J Psychiatr Res 1975; 12:189-198.

6. Yesavage JA, Brink TS, Rose TL, et al. Development and

validation of a geriatric depression rating scale: a pre-
liminary report. J Psychiatr Res 1983; 17:37-49.

7. Zarit SH, Reever KE, Bach-Peterson J. Relatives of the

impaired elderly: correlates of feelings of burden. Ger-
ontologist 1980; 20:649-655.

8. Health and Public Policy Committee, American College of

Physicians. Home health care. Ann Intern Med 1986;
105:454-^60.

9. American Geriatrics Society Public Policy Committee.

Home care and home care reimbursement. J Am Geri-
atr Soc 1989; 37:1065-1066.

10. Council on Scientific Affairs, American Medical Associa-
tion: Educating physicians in home health care. JAMA
1991; 265:769-771.

Community Initiatives to Improve the
Lives of Older Adults in East Harlem

Judith L. Howe, M.A., M.P.A.

The Department of Geriatrics and Adult De-
velopment was established in 1982 at the Mount
Sinai Medical Center in recognition of the need
for an increased focus on the clinical, research,
educational, and policy needs of a rapidly aging
population. The overall goals of the department
are to provide hospital-based and community-fo-
cused comprehensive health care; to conduct ed-
ucation and training for physicians, other health
professionals, and the public; and to carry out ag-
ing-related research and policy studies.

Mount Sinai serves a broad catchment area,
including East Harlem, which is an inner-city
multiracial community comprised of lower-in-
come and working-class residents. The elderly of
East Harlem are particularly needy in a number
of ways, including health care, social services,
and housing. Since its founding, the department
has actively reached out to the East Harlem el-
derly through a variety of approaches.

Background

As providers of a range of health care and
other services to the older population of East
Harlem, the department is challenged to work
with its clients to achieve an optimal quality of
life in the context of numerous difficulties. East
Harlem's elderly face a multitude of problems as
a result of financial hardship, shifting family and
social support patterns, and inadequate access to
needed services. Chief among the problems are
inadequate living situations and incomes and dif-
ficulty in accessing health and social services. In

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, New York, New York. Ad-
dress reprint requests to J. Howe, Department of Geriatrics,
Box 1070, The Mount Sinai Medical Center, New York, NY
10029.

492

addition, many older people have no involvemem
in family and community life.

East Harlem comprises the northeastern sec
tion of the borough of Manhattan, bounded on th(
south by East 96th Street, on the west by Fiftl
Avenue, on the east and north by the East anc
Harlem rivers. The total population of Easi
Harlem is 108,408, with 11.6%, or 12,615, aged 61
or older in 1990, an increase from 10.5% in 1980
The ethnic composition of East Harlem is mixed
According to 1990 census data, 38.7% of the tota
population is non-Hispanic black and 52.3% ii
Hispanic. Residents who speak little or no En
glish constitute almost 13% of the population (1)

Thirty-four percent of those 65 and older liv
ing in East Harlem are below the poverty lin(
according to 1990 census figures. In 1986 the me
dian household income in East Harlem ($8,305
was only 41.5% of the city's median ($20,000). Foi
older residents living in housing owned and man
aged by New York City, the median income was
even lower, those on public assistance having i
1986 median income of only $4,428 (1).

The health status of the East Harlem elderlj
is worse than that of other older people in Nev
York City, as measured by higher mortalitj
rates. There is excess mortality resulting frorr
conditions that are greatly influenced by th(
physical and social environment and inadequate
access to health care, such as pneumonia and in
fluenza, diabetes, and homicide. Also, comparec
to other older people in New York City, most Easi
Harlem elderly have chronic conditions thai
make using public transportation difficult or im-
possible for them, thus increasing their isolatior
(2).

A 1989 report on primary care in Easi
Harlem (3) found that the primary-care needs
most lacking for older people are mental health
services, home care and transportation services,

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

PROGRAMS FOR EAST HARLEM ELDERLY— HOWE

493

and coordination of health care. The report rec-
ommended that steps be taken to integrate phys-
ical and mental health services with other ser-
vices such as day care, home care, transportation,
housing, senior centers, settlement houses, and
religious institutions.

Departmental Community
Outreach Programs

Since its establishment in 1982, the Depart-
ment of Geriatrics has developed a number of pro-
grams geared to serving the needs of the older
residents of East Harlem:

The Phyllis and Lee Coffey Outpatient Geri-
atrics Clinic serves approximately 700 patients in
continuing primary care, most of w^hom live in
East Harlem, Central Harlem, and other nearby
communities.

The Geriatrics Home Medical Care Program
provides comprehensive health services and man-
agement to homebound older people in East
Harlem who have no other access to medical care.

The Well Elderly Program, in conjunction
with the Senior Division of the 92nd Street Young
Men's- Young Women's Hebrew Association, pro-
vides year-round health promotion and disease
prevention services.

The Elder Abuse Assessment and Assistance
Program provides direct service to individuals
and families, as well as consultation, training,
and resource materials to health care profession-
als at The Mount Sinai Hospital and in the com-
munity.

The Hunter College-Mount Sinai Geriatric
Education Center offers ongoing training pro-
grams for community health professionals in the
prevention, treatment, and diagnosis of disease.

The Alzheimer's Disease Research Center
(ADRC), jointly sponsored by the Departments of
Psychiatry and Geriatrics at Mount Sinai Medi-
cal Center, offers community education and out-
reach programs and support groups for patients,
family members, and caregivers. Recently, the
ADRC began working with a group of East
Harlem senior centers to establish an educational
program for seniors and staff regarding Alzhei-
mer's disease. This program will also assess the
prevalence of Alzheimer's disease among the His-
panic elderly in East Harlem, including follow-up
treatment at Mount Sinai when indicated.

Project Linkage

In response to the department's concern for a
comprehensive approach to the multitude of prob-
lems facing not only the elderly, but East Harlem
residents of all ages, funding was sought to assist

the department in assessing the feasibility of de-
veloping shared housing for older people in East
Harlem. A unique model was envisioned, one
which incorporated housing, after-school care for
latchkey children, intergenerational program-
ming, and health services.

The objectives of the program, called Project
Linkage, are:

To increase the availability of affordable
housing for low-income, elderly residents of East
Harlem

To expand the availability of after-school
care for East Harlem children aged 5 through 12

To provide meaningful roles for older people
through intergenerational programs and activi-
ties shared with residents of the project

To improve access to health care services

In 1991 the department received a one-year
grant from The New York Community Trust for a
study to assess the feasibility of developing
shared housing for the elderly, coupled with after-
school programs for children, intergenerational
programming, and health care services. A study
aided by the Dept. of Community Relations, a
consultant team, a projects advisory committee,
and focus groups of East Harlem elderly residents
found Project Linkage feasible.

In early 1993 a Mount Sinai team solicited
capability statements from East Harlem commu-
nity-based organizations interested in serving as
cosponsors with Mount Sinai in developing and
operating Project Linkage. After a systematic re-
view process, a partnership was formed with the
Community Association of the East Harlem Tri-
angle, the Union Settlement Association, and the
Greater Emmanuel Baptist Church to sponsor
housing for the elderly under the HUD section
202 program. The respective boards of directors of
all organizations endorsed the project and agreed
to the conditions of sponsorship. The manage-
ment team, consisting of representatives from the
participating organizations, as well as a housing
consultant and an architect, selected a site on
East 118th Street between First and Second Av-
enues and developed an application for HUD sec-
tion 202 financing.

Certain elements of the original model were
modified to conform to strict HUD guidelines on
design features and financing. For instance,
whereas the feasibility study proposed a building
of 50 studio apartments, the planned building has
70 one-bedroom apartments because of economies
of scale.

Project Linkage is based on a shared housing
model designed to facilitate communal living and

494

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

the formation of informal support networks,
while allowing for residents' privacy. Each apart-
ment will have its own living room and kitchen-
ette; in addition, each cluster of five or six apart-
ments will have shared living and dining space to
promote interaction among residents. There will
also be common space for intergenerational activ-
ities, after-school programs, social programs, and
health visits and screenings.

One strength of the model is the availability
of on-site health and mental health services pro-
vided by the Department of Geriatrics and the
Union Settlement Association. A consultation
and examination room will be available on the
first floor for health screenings and visits. Ini-
tially, residents will be healthy and independent,
but the model will encompass "aging in place" as
residents grow older. Successful aging in place is
dependent on such factors as the availability of
assistance from an informal support system and
access to formal health and social services. Proj-
ect Linkage meets both these requirements: the
housing arrangement is such that the formation
of natural helping networks is encouraged, and
residents as they age will continue to have access
to the continuum of health and social services
provided by Mount Sinai and Union Settlement.

The full range of medical services along the
care continuum from independence to dependence
offered by the Department of Geriatrics and
Adult Development and The Mount Sinai Medical
Center will support Project Linkage residents as
they age. These include:

• Health education and health promotion pro-
grams

• Specialized outpatient primary care

• Referral to the many specialized clinical ser-
vices at The Mount Sinai Hospital as needed
(for instance, rehabilitation, radiology, osteopo-
rosis, and geropsychiatry)

• Admission to The Mount Sinai Hospital spe-
cialized geriatrics inpatient unit, which focuses
on the treatment of frail older persons

• For those East Harlem residents admitted to
other units at The Mount Sinai Hospital, the
services of the Geriatrics Consultation Service

• The Geriatrics Home Medical Care Program for
those older individuals unable to travel to the
hospital for care

In early October 1993 The Mount Sinai Hos-
pital was notified by HUD that it had received a
grant of almost $5.8 million to proceed with build-

ing section 202 housing for Project Linkage at the
selected site.

Conclusions

The department has provided a range of
health services to the East Harlem elderly during
its first ten years. In addition to direct delivery
of service, it has played an important role in de-
veloping community-based programs in East
Harlem.

Working coalitions between large health care
institutions and community organizations are of-
ten difficult to develop and sustain. For instance,
in 1986 the Geriatrics and the Community Med-
icine departments spearheaded the formation of a
coalition of about 13 East Harlem agencies to co-
ordinate health and social services for frail older
people, called the East Harlem Living at Home
Program. The coalition, however, struggled to
meet the program goal of targeting older people
at risk for institutionalization and providing co-
ordinated services, in large part because of staff-
ing problems and difficulties in maintaining the
coalition structure.

In Project Linkage, great care was taken to
build a coalition made up of committed and strong
member organizations, with Mount Sinai,
through the departments of Geriatrics and Com-
munity Relations, playing a more active organi-
zational role in program development and man-
agement.

Acknowledgments

The author acknowledges the financial support of The New
York Community Trust and United Jewish Appeal/Federation
of Jewish Philanthropies for technical assistance in planning
and developing Project Linkage. In addition, the following in-
dividuals have played key roles in Project Linkage, and the
author acknowledges their contributions: Robert N. Butler,
M.D., Gary Rosenberg, Ph.D., Barbara Brenner, Dr.P.H.,
Glenn Williams, Helene Clark, Richard Silverblatt, Joanne
Hoffman, and Maria Gutierrez.

References

1. Final report of Project Linkage feasibility study and pre-

sentation of a model for shared housing for the elderly
and intergenerational programming in East Harlem,
prepared for Department of Geriatrics and Adult De-
velopment, The Mount Sinai Medical Center, August
1992.

2. The elderly of East Harlem: people, resources and service

needs, a report of the Department of Community Med-
icine, Mount Sinai School of Medicine, 1987.

3. Report on primary care in East Harlem: a local strategy,

a report of the East Harlem [New York, NY] Commu-
nity Health Committee, 1989.

Education, Clinical Services, and Research
On Treating Older People

Management of Geriatric Disorders

Practical Pharmacology for
Older Patients:

Avoiding Adverse Drug Effects

Judith Ahronheim, M.D.

Older individuals take larger numbers of pre-
scription drugs than do their younger counter-
parts (1, 2). According to national databases, the
number of adverse drug events (ADEs) increases
with age (2). This is partly (4, 5), though not en-
tirely, related to increased drug exposure, for the
older the patient, the more likely the ADE will be
serious (3). Principles of geriatric pharmacology
have been widely reviewed (&-8), and, to pre-
scribe appropriately, the physician should keep
these principles in mind. This paper highlights
various factors that contribute to the geriatric pa-
tient's increased susceptibility to ADEs.

Altered Pharmacokinetics

With advancing age come alterations in
pharmacokinetics — that is, the impact of distri-
bution, metabolism, and elimination of drugs on
the timing of a drug's effects.

Renal Function. The most consistently de-
scribed change is due to an age-related decline in
renal function. On the average, glomerular filtra-
tion rate declines approximately 50% between the
ages of 20 and 70 years (9, 10). This degree of
renal functional decline from a normal baseline
does not produce illness; however, it may be in-
adequate for the elimination of active substances
given exogenously, and toxic levels of drugs may
occur if doses are not properly adjusted. This is
especially important when the drug has a narrow

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, New York, NY. Address
reprint requests to: Judith Ahronheim, M.D., Department of
Geriatrics and Adult Development, Box 1070, One Gustave L.
Levy Place, The Mount Sinai Medical Center, New York, NY
10029.

therapeutic index (therapeutic-to-toxic ratio). Re-
nally eliminated penicillin has a wide therapeutic
index and only rarely produces dose-related tox-
icity, so age-adjusted dosing is considered op-
tional. In contrast, aminoglycoside antibiotics,
such as gentamicin, produce nephrotoxicity when
only slightly supratherapeutic doses are given. It
is thus essential that the dose of the latter group
of drugs be carefully adjusted according to the
patient's renal function. Other important renally
excreted drugs with a narrow therapeutic index
include digoxin, vancomycin, and imipenem. It
has been our experience on the Mount Sinai Ge-
riatrics Consultation Service that the most com-
mon prescribing error on medical and nonmedical
wards is the overdosing of renally excreted anti-
biotics with a narrow therapeutic index.

Not all geriatric patients have significant de-
clines in glomerular filtration rate. In fact, longi-
tudinal studies have shown that at least one-third
of the elderly maintain stable renal function for a
period of years (11). Thus, age alone is not predic-
tive of the degree of functional renal decline, and
some other predictor must be used. The clinician
is most likely to measure renal function using
standard laboratory tests, such as creatinine
clearance or serum creatinine. However, because
muscle mass declines with age, serum creatinine,
a reflection of muscle mass, may fail to reflect a
significant age-related decrease in creatinine
clearance. Under the best of circumstances, it is
cumbersome to collect an accurate 24-hr urine
specimen for creatinine clearance determination;
for many elderly patients, collecting such a spec-
imen may prove impossible in practice. A popular
formula used to estimate creatinine clearance is
(12):

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

497

498

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Creatinine clearance -

weight (kg) X (140 - age)
72 X serum creatinine (mg/dL)
For women, multiply result x 0.85

Because of the wide variation in renal func-
tion among the elderly, this formula may both
overestimate and underestimate actual renal
function. Since such a formula assumes that renal
function is stable, it cannot be relied upon in the
acutely ill, whose renal function may change
daily. Furthermore, in the debilitated elderly,
this formula often overestimates actual creati-
nine clearance by as much as 20% (13), presum-
ably because the correlation between lean body
weight and muscle mass is altered. However, pre-
dictive formulas have practical applicability in
certain situations. For example, in an acutely ill
hospitalized patient who requires a drug with a
narrow therapeutic index, it is helpful to calcu-
late the dosage based on estimated renal function
until serum drug levels can be obtained and the
dosage can be readjusted accordingly.

Loading Dose and Maintenance Dose. Note
that the loading dose for drugs does not necessar-
ily have to be age-adjusted. Although the phar-
macodynamic effect of many drugs may be en-
hanced in late life (see below), toxicity based on
altered pharmacokinetics develops only after re-
peated dosing. When a therapeutic effect is ur-
gently needed, as in serious infection, adequate
serum and tissue levels must be attained. This is
achieved if a non-age-adjusted loading dose is
given, and dose-related toxicity is avoided if the
maintenance dose is tailored to the patient's abil-
ity to eliminate the drug.

Hepatic Function: Oxidation and Conjuga-
tion. Certain changes in hepatic function also oc-
cur with advancing age. Those that affect drug
handling include decreased hepatic blood flow
(14) and altered activity of oxidative enzymes in-
volved in drug metabolism (15). Decreased he-
patic blood flow reduces the efficiency with which
the liver can handle drugs that present for me-
tabolism. However, there is no unanimity of opin-
ion regarding age-related hepatic function (15-
17), and there are no simple measures of these
changes in hepatic processes. Furthermore, it has
been difficult for pharmacokinetic studies to
home in on individual hepatic processes involved
in drug handling, since drug metabolism depends
on the coordination of complex processes in a
number of organ systems. In general, it has been
shown that hepatic oxidation ("phase I metabo-
lism") may decline somewhat with age, whereas
hepatic conjugation ("phase II metabolism") does
not.

Some oxidated metabolites are pharmacolog-
ically active, and they are also excreted renally.
The combined effect of this is that the action of
many hepatically oxidized drugs can be markedly
prolonged in older individuals. For example, flu-
razepam (Dalmane), a benzodiazepine hypnotic,
is oxidized to dealkylflurazepam, a renally ex-
creted metabolite that is pharmacologically ac-
tive. Owing to the activity of this metabolite, the
half-life of one 15-mg dose of flurazepam may be
twice as long in the overtly healthy elderly as in
younger adults, averaging about 120 hr in elderly
women and 160 hr in elderly men (18).

The cytochrome P-450 system, the major
source of hepatic oxidative enzymes, is affected by
exogenous substances, including a wide variety of
drugs. These substances may either inhibit or
stimulate the enzyme system, producing an alter-
ation in the drug's duration of action. There is
controversy over whether these modifications (en-
zyme inducibility or inhibition) are altered in late
life. However, enzyme inducibility and inhibition
is responsible for many drug-drug interactions,
and the elderly patient's increased exposure to
drugs increases the likelihood of such interac-
tions (5).

Nonoxidative Processes. Many drugs are me-
tabolized in the liver by conjugation with a new
molecule. These nonoxidative processes are not
significantly altered with age; furthermore, the
resulting metabolites tend to be inactive. Thus,
drugs metabolized by hepatic conjugation do not
generally have an increased half-life in elderly
patients. This is the case for oxazepam and lora-
zepam, which are themselves active metabolic
products of other benzodiazepines.

Volume of Distribution. The volume of dis-
tribution (Vd) for drugs may be altered with age.
The Vd is the hypothetical space in which the
total amount of drug in the body would distribute
to achieve the same concentration in plasma. The
Vd is calculated by dividing the total amount of
drug by the concentration that amount has
achieved in plasma. The ratio of fat-to-lean body
weight increases with age, more in women than
in men. As a consequence, the Vd for fat-soluble
drugs increases with age, and steady-state con-
centrations for lipid-soluble drugs, such as the
benzodiazepines, may be reached later than ex-
pected. If the patient continues to take a drug like
flurazepam, which has an increased Vd (in addi-
tion to slowed oxidation and a renally excreted
active metabolite), or increases the dose, delayed
toxicity may occur. The patient or physician may
fail to recognize that an adverse effect is due to
the drug, which might have been initiated two
weeks earlier. Conversely, the volume of distri-

Vol. 60 No. 6

PRACTICAL GERIATRIC PHARMACOLOGY— AHRONHEIM

499

bution for water-soluble drugs tends to decrease
with age. As a consequence, the peak effect of a
substance like alcohol, which distributes in wa-
ter, tends to occur earlier than expected.

Protein Synthesis. Protein synthesis is im- •
paired with age (19). Serum levels of albumin
tend to remain in the normal range among
healthy elderly persons, but a number of acute
and chronic illnesses may rapidly produce hypoal-
buminemia (20), and this condition is probably
more likely to occur in the elderly patient with
diminished protein intake (21). This may have
implications for prescribing drugs that are highly
protein bound. If a drug is highly bound to albu-
min, a decrease in albumin leads to an increase in
the proportion of drug that is unbound (free), the
active fraction of the drug. Free drug travels more
rapidly to the liver and kidney than bound drug
and, therefore, is also more swiftly eliminated;
thus, in the presence of hypoalbuminemia, en-
hanced activity of the drug is unlikely to last a
significant amount of time, if at all. However, the
development of hypoalbuminemia may become
important when one tries to interpret serum lev-
els of drugs highly bound to albumin (22). Rou-
tine clinical laboratory determinations measure
total drug, which includes both the bound and
free fractions. Therefore, a patient who is hypoal-
buminemic may have a subtherapeutic to normal
serum level of total drug, but a normal to supra-
therapeutic level of free drug. In such a situation,
the clinician might mistakenly increase the dose
of the drug based on the reported level of total
drug. If a highly bound drug also has a narrow
therapeutic index, as does, for example, pheny-
toin, a slight dose increase could lead to unex-
pected toxicity.

A number of drugs have affinity not for albu-
min but for alpha-1 acid glycoprotein (AAG). This
protein is an acute-phase reactant and, in con-
trast to albumin, tends to increase rather than
decrease with age and illness. AAG is not deter-
mined by routine laboratory studies, fluctuates
greatly, and therefore plays a limited role in clin-
ical decision making. However, knowledge of
AAG may help to explain peculiarities of dose and
effect. For example, following a myocardial in-
farction, AAG levels tend to increase, leading to
an increase in total serum levels of lidocaine, a
drug commonly used in the peri-infarction period.
Because lidocaine is extensively bound to AAG,
free lidocaine levels would decrease, so total se-
rum levels might overestimate the potential clin-
ical effect of the drug (23). There is insufficient
information on the degree to which free levels cor-
relate with clinical effect, and this interesting
phenomenon awaits further study. However, it is

important to recognize this phenomenon when in-
terpreting total levels of highly bound drugs in
the elderly.

Altered Pharmacodynamics

Pharmacodynamics refers to the specific ac-
tion of the drug at the tissue level. The pharma-
codynamic effect of a given drug may be enhanced
or decreased in the elderly, owing to physiologic
and pathologic changes in both target and non-
target organs. For example, beta-receptor sensi-
tivity may be reduced in late life, leading to di-
minished response to both beta agonists and
antagonists (24). Although it has been reported
that elderly patients achieve less hypotensive re-
sponse to beta blockers than do younger adults
(25), another study revealed that metoprolol was
more effective among elderly than among many
younger cardiac patients in preventing recurrent
myocardial infarction (26). Both inadequate hy-
potensive response and enhanced cardioprotec-
tion could perhaps be explained by other factors,
however, so the clinical implications of impaired
beta-receptor sensitivity are not known.

More commonly, pharmacodynamic effects of
drugs are enhanced — or appear to be enhanced —
in older patients. Important examples are drugs
that affect the central nervous system, which are
a common cause of confusional states in the el-
derly (4). This may be due to the high prevalence
of underlying Alzheimer's disease, other central
nervous system lesions, or age- and disease-re-
lated alterations in neurotransmitter concentra-
tions. The patient may be cognitively normal un-
til exposed to the ofi'ending agent because the
potential central nervous system deficits are sub-
clinical. Thus, the clinician should be alert for the
development of confusional states in the elderly
when a variety of drugs are given.

Drugs that can cause confusion include:

• Any sedative

• Any hypnotic

• Antidepressants

• Anticonvulsants

• Centrally acting antihypertensives

• Lidocaine

• Digoxin

• Isoniazid

• Corticosteroids

• Theophylline

• Anticholinergic agents

• Nonsteroidal anti-inflammatory agents

• Histamine-2 blockers

• Any drug until proven otherwise

It is not always possible to determine when
an enhanced effect is related to altered pharma-

500

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

TABLE

Pharmacodynamic Alterations in the Elderly

Possible
outcome
or side

Drug effect

Alteration

Impaired baroreceptor

Diuretics;

Orthostatic

function; venous

tricyclics;

hypotension

insufficiency

methyldopa;

others

Impaired counter-

Hydralazine

No reflex

regulatory mechanisms

tachycardia

Altered temperature

Phenothiazines

Hypo- or

regulation

hyperthermia

Increased ADH

Chlorpropamide;

Hyponatremia

secretion

others

Decreased androgenic

Digoxin;

Gynecomastia

hormones (males)

spironolactone

Underlying disorder*

Atrophic gastritis

Aspirin;

Gastric

NSAIDs

hemorrhage

Immobility; cathartic

Many

Constipation

bowel

Dementia

Many

Confusional

state

Sinus or AV nodal

Digoxin;

Bradycardia

disease

verapamil

Venous insufficiency

Nifedipine

Edema

Unstable bladder

Diuretic

Incontinence

Prostatic hyperplasia

Anticholinergics;

Urinary

tricyclics;

retention

disopyramide

Parkinson's disease

Metoclopramide;

Parkinsonian

neuroleptics

symptoms

* May be subclinical.

Adapted with permission from (27).

codynamics or pharmacokinetics. To prove that
there is enhanced sensitivity at the tissue level,
one would have to demonstrate an enhanced ef-
fect in the face of an ordinary serum or, more
precisely, tissue concentration of the drug. In
many cases, the patient's seeming overreaction
might well be related to supratherapeutic levels
of the drug caused by diminished elimination.
However, clinical experience has shown that ad-
verse drug events often manifest themselves in
specific ways in the elderly patient. Examples of
drugs and their potential effects, based on appar-
ent altered pharmacodynamics, are listed in the
Table.

Summary

Older patients experience adverse drug
events more frequently than do younger adults,
and they experience them in unexpected ways.
The heterogeneity of the geriatric population
leaves the investigator with a great deal of un-
certainty and the clinician with little reliable
guidance based on clinical investigations. Al-
though few generalizations can be made, it is safe
to say that at least a subgroup of patients exhibit
one or more physiologic or pathologic alterations

that can lead to unanticipated drug response. It is
essential for the prescribing clinician to be mind-
ful of the vulnerability of this population. Pre-
scribing should be judicious and, in general, con-
servative, with an eye on the patient's specific
vulnerabilities.

References

1. Darnell JC, Murray MD, Martz BL, Weinberger M. Med-

ication use by ambulatory elderly. An in-home survey.
J Am Geriatr Soc 1986; 34:1^.

2. Nolan L, O'Malley K. Prescribing for the elderly. Part II.

Prescribing patterns: differences due to age. J Am Geri-
atr Soc 1988; 36:245-254.

3. Burke LB, Jolson HM, Goetsch RA, Ahronheim JC. Geri-

atric drug use and adverse drug event reporting in
1990: a descriptive analysis of two national data bases.
Annu Rev Gerontol Geriatr 1992; 12:1-28.

4. Larson EB, KukuU WA, Buchner D, Reifler BV. Adverse

drug reactions associated with global cognitive impair-
ment in elderly persons. Ann Intern Med 1987; 107:
169-173.

5. May FE, Steward RB, Cliff LE. Drug interactions and

multiple drug administration. Clin Pharmacol Ther
1977; 22:322-328.

6. Swift CG, Triggs EJ. Clinical pharmacokinetics in the el-

derly. In: Swift CG, ed. Clinical pharmacology in the
elderly. New York: Marcel Dekker, 1987:31-82.

7. Feely J, Coakley D. Altered pharmacodynamics in the el-

derly. Clin Geriatr Med 1990; 6:269-283.

8. Dawling S, Crome P. Clinical pharmacokinetic consider-

ations in the elderly. Clin Pharmacokinet 1989; 17:
236-263.

9. Davies DF, Shock NW. Age changes in glomerular filtra-

tion rate: effective renal plasma flow and tubular ex-
cretory capacity in adult males. J Clin Invest 1950;
29:496-507.

10. Rowe JW, Andres R, Tobin J, et al. The effect of age in

creatinine clearance in men: a cross sectional and lon-
gitudinal study. J Gerontol 1976; 31:155-163.

11. Lindeman RD, Tobin J, Shock NW. Longitudinal studies

on the rate of decline in renal function with age. J Am
Geriatr Soc 1985; 33:278-285.

12. Cockcroft DW, Gault MH. Prediction of creatinine clear-

ance from serum creatinine. Nephron 1976; 16:31-41.

13. Drusano GL, Muncie HL, Hoopes JM, Damron DJ, War-

ren JW. Commonly used methods of estimating creati-
nine clearance are inadequate for elderly debilitated
nursing home patients. J Am Geriatr Soc 1988; 36:437-
441.

14. Woodhouse KW, Wynne HA. Age-related changes in liver

size and hepatic blood flow. The influence on drug me-
tabolism in the elderly. Clin Pharmacokinet 1988; 15:
287-294.

15. Vestal RE, Norris AH, Tobin JD, Cohen BH, Shock NW,

Andres R. Antipyrine metabolism in man: Influence of
age, alcohol, caffeine and smoking. Clin Pharmacol
Ther 1975; 18:425-432.

16. Greenblatt DJ, Divoll MK, Harmatz JS, Shader RI. An-

tipyrine absorption and disposition in the elderly. Phar-
macology 1988; 36:125-133.

17. Woodhouse KW, Mutch E, Williams FM, Rawlins MD,

James OFW. The effect of age on pathways of drug me-
tabolism in human liver. Age Ageing 1984; 13:328-334.

18. Greenblatt DJ, Divoll M, Harmatz JS, MacLaughlin DS,

Vol. 60 No. 6

PRACTICAL GERIATRIC PHARMACOLOGY— AHRONHEIM

501

Shader RI. Kinetics and clinical effects of flurazepam in
young and elderly noninsomniacs. Clin Pharmacol Ther
1981; 30:475-486.

19. Banner DB, Holbrook NJ. Alteration in gene expression

with aging. In: Schneider EL, Rowe JW, eds. Handbook
of the biology of aging, 3rd ed. San Diego: Academic
Press, 1990:97-115.

20. Campion EW, deLabry LO, Glynn RJ. The effect of age on

serum albumin in healthy males: report from the nor-
mative aging study. J Gerontol 1988; 43:M18-M20.

21. Young VR. Amino acids and proteins in relation to the

nutrition of elderly people. Age Ageing 1990; 10:S10-
S24.

22. Greenblatt DJ, Sellers EM, Koch-Weser J. Importance of

protein binding for the interpretation of serum or
plasma drug concentrations. J Clin Pharmacol 1982;
22:259-263.

23. Routledge PA, Stargel WW, Wanger GS, Shand DG. In-

creased alpha-l-acid glycoprotein and lidocaine dispo-
sition in myocardial infarction. Ann Intern Med 1980;
93:701-704.

24. Vestal RE, Wood A J J, Shand DG. Reduced beta-adreno-

ceptor sensitivity in the elderly. Clin Pharmacol Ther
1979; 26:181-186.

25. Buhler F^R, Burkart F, Lutold BE, Kung M, Marbet G,

Pfisterer M. Antihypertensive beta blocking actions as
related to renin and age: a pharmacologic tool to iden-
tify pathogenic mechanisms in essential hypertension.
Am J Cardiol 1975; 36:653-669.

26. Olsson G, Rehnqvist N, Sjogren A, Erhardt L, Lundman

T. Long-term treatment with metoprolol after myocar-
dial infarction: effect on 3 year mortality and morbid-
ity. J Am Coll Cardiol 1985; 5:1428-1437.

27. Ahronheim JC. Handbook of prescribing medications for

geriatric patients. Boston: Little, Brown, 1992:3.

Evaluation and Management of Urinary
Incontinence in Older Patients

Perry Starer, M.D.

Urinary incontinence occurs in 10% to 43% of
elderly individuals aged 65 years or older living
in the community (1—6) and 36% to 62% of elderly
patients living in long-term-care facilities (7-10).
Complications from urinary incontinence include
skin breakdown, infection, and loss of self-esteem
(11). The economic cost of urinary incontinence is
high because of its many ramifications, including
laundry, disposable diapers, medical complica-
tions, and institutionalization (12, 13). The social
and emotional cost is also high, though more dif-
ficult to quantify, contributing to isolation, em-
barrassment, and frequent institutionalization
(13). Successful treatment of incontinence may
provide many older individuals with an improved
quality of life and an opportunity to remain inde-
pendent in the community.

Management of urinary incontinence per-
plexes and distresses both patient and clinician.
Appropriate diagnostic and therapeutic interven-
tions are often overlooked. Instead, diapers or
catheters are applied and the problem is con-
cealed (14). To select the appropriate treatment
for an elderly incontinent patient, the cause of the
urinary incontinence needs to be determined.

Patterns of Abnormality

Urinary incontinence is the initial manifes-
tation of a variety of underlying pathologic pro-
cesses. Incontinence can be classified by the fol-
lowing patterns of presentation: detrusor

From The Jewish Home and Hospital for Aged and the De-
partment of Geriatrics and Adult Development, Mount Sinai
School of Medicine, New York, NY. Address reprint requests
to Perry Starer, M.D., Department of Medicine, The Jewish
Home and Hospital for Aged, 120 West 106th Street, New
York, NY 10025.

instability; overflow incontinence: urinary reten-
tion and detrusor-urethral sphincter dyssynergia;
stress incontinence: sphincter insufficiency; func-
tional incontinence; and iatrogenic incontinence
(14). The most common urodynamic abnormality
in incontinent geriatric outpatients (15), detrusor
instability (with or without intact detrusor con-
tractility), is also the most common cause of po-
tentially correctable urinary incontinence in
nursing home patients (16). However, there may
be multiple contributory factors (17).

Neurourology. The central nervous system
receives afferent impulses from stretch receptors
in the bladder wall during filling (18). The con-
cept of a "cortical inhibitory area acting on a sac-
ral reflex arc" may be an oversimplification (19).
The sacral micturition reflex is influenced by fa-
cilitatory and inhibitory areas located at different
levels of the neural axis (19). Diencephalic and
brain stem areas controlling the micturition re-
flex include a cerebral inhibitory region, a hypo-
thalamic facilitatory area in the mammillary re-
gion of the posterior hypothalamus, a midbrain
inhibitory area localized bilaterally in the
tegmentum lateral to the central gray matter at
the caudal superior collicular level, and a pontine
facilitatory area localized bilaterally in the dorsal
tegmentum at the isthmus level immediately
ventral to the lateral angles of the periventricu-
lar gray matter (19, 20).

Detrusor Instability. Detrusor instability,
the occurrence of uninhibited detrusor contrac-
tions (14), is a common urodynamic finding
among elderly incontinent patients (8, 21-24). In-
voluntary, uninhibited detrusor contractions may
be produced either by increased afferent impulses
to the central nervous system due to local disor-
ders in the bladder or the urethra or by a neuro-
logic disorder affecting the inhibitory nervous

502

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

URINARY INCONTINENCE IN THE ELDERLY— ST ARER

503

pathways from cortical and subcortical centers
(18). Provocative upright cystometry is helpful in
diagnosing involuntary contractions not pro-
voked in the supine position (25). The symptoms
of urinary frequency and urgency may occur in
the presence of detrusor instability without being
associated with urinary incontinence (12).

Detrusor instability may be associated with
bladder outlet obstruction (26-28), but there is no
correlation between the severity of obstruction
and the presence of instability (29, 30). Detrusor
instability is reversible in the majority of patients
undergoing surgery for bladder outlet obstruction
(28, 30). Detrusor instability may also be found in
women with stress incontinence (15, 31). Since
detrusor instability has been reported as being
present in nonincontinent subjects (15, 32, 33),
the clinical significance of this entity becomes un-
clear. An elderly patient with involuntary blad-
der contractions may be continent (32, 33), but
following a cerebrovascular accident, the combi-
nation of impaired mobility and involuntary con-
tractions may lead to urinary dysfunction.

Overflow Incontinence: Urinary Retention.
Urinary retention may occur secondary to blad-
der outlet obstruction or poor bladder contrac-
tions. Medications may cause urinary reten-
tion, and anticholinergic agents can inhibit
bladder contractions. Alpha-adrenergic stimulat-
ing agents can increase outlet and urethral pres-
sure. A patient's inability to voluntarily contract
the bladder during cystometric testing may be a
result of discomfort during the procedure, leading
to an incorrect diagnosis of a noncontractile blad-
der. Normal patients may be unable to produce a
bladder contraction during cystometric studies,
even though no abnormality exists (34). An in-
ability to urinate in the presence of other people
could be mistaken for bladder dysfunction.

The choice of therapy depends on the cause of
retention. Surgical treatment may be indicated in
the setting of bladder outflow obstruction. Most
patients experience an improvement in prostatic
symptoms following transurethral resection of
the prostate for benign prostatic hypertrophy
(35). Since the symptomatology of detrusor insta-
bility without infravesical obstruction is similar
to the symptomatology of prostatism, it is impor-
tant to obtain objective evidence of infravesical
obstruction before surgery (30, 36, 37). It is diffi-
cult to use the symptoms of prostatism as an in-
dex of whether obstruction has been relieved be-
cause symptomatic improvement postoperatively
may be related to a decrease in detrusor instabil-
ity (36). The results of prostatectomy are poor

when done in patients within a year following a
cerebrovascular accident (38). Postoperative uri-
nary incontinence in patients with cerebrovascu-
lar disease may be due to detrusor dysfunction or
reduced cerebral control of micturition (39).

Even though bethanechol chloride, a cholin-
ergic agent, is pharmacologically active, it has
not been shown to be effective in promoting blad-
der emptying (40). Crede's maneuver (the appli-
cation of steady suprapubic pressure over the full
bladder in a caudal direction) is an inefficient
method of bladder emptying for most patients
(41). Although prostatectomy is the currently ac-
cepted treatment of benign prostatic hyperplasia,
the role of pharmacologic therapy is being inves-
tigated (42). The clinical efficacy of alpha-adren-
ergic blockers is related to relaxation of the
smooth muscle component of the prostate (42).
The alpha-adrenergic blocker phenoxybenzamine
may have a place in the management of patients
with symptoms attributable to bladder outflow
obstruction, especially when surgical treatment
needs to be delayed (43, 44). Terazosin, a long-
acting selective alpha- 1 blocker, is a safe and ef-
fective therapy for symptomatic benign prostatic
hyperplasia (45). Treatment of benign prostatic
hyperplasia with finasteride, a competitive inhib-
itor of 5-alpha-reductase, results in decreased se-
rum dihydrotestosterone concentrations, a de-
crease in prostatic volume, and an increase in
urinary flow (46).

Overflow Incontinence: Detrusor-Urethral
Sphincter Dyssynergia. Detrusor-extemal sphinc-
ter dyssynergia is the inappropriate contraction
or failure of relaxation of the external urethral
sphincter during detrusor contractions (47, 48).
This discordant micturition reflex can elevate
the residual urine (49). The normal relationship
between the detrusor and the external sphincter
is disrupted after spinal cord injury interrupts
the pathways connecting the pontine mesen-
cephalic and the sacral micturition centers (48,
50).

Some investigators have hypothesized that
detrusor-sphincter dyssynergia is an abnormal
flexor response of the perineal musculature to
bladder contraction (51), whereas others have
proposed that external sphincter dyssynergia is
an exaggerated continence reflex (increasing ex-
ternal urethral sphincter electromyographic ac-
tivity with bladder filling) (52). The loss of conti-
nuity between the sacral cord pudendal and
vesical motor nuclei and the supraspinal micturi-
tion centers results in the loss of appropriate in-
hibition of the pudendal motor nucleus during a

504

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

detrusor contraction (52). Beclofen, an antispastic
drug, decreases external urethral sphincter resis-
tance and reduces residual urine in with patients
spinal cord injury (49).

Stress Incontinence: Sphincter Insuffi-
ciency. Stress urinary incontinence is the invol-
untary loss of urine caused by an increase in in-
tra-abdominal pressure but not by contraction of
the detrusor muscle (53). It has been described as
the most common cause of urinary incontinence
in women (53). Although a surgical procedure is
often recommended as treatment, anatomic stress
incontinence is not present in all incontinent
women, who, in turn, may not benefit from a sur-
gical procedure (54). Before surgical therapy is
recommended for incontinent women, it is impor-
tant to determine the etiology of their urinary
dysfunction.

Although the evaluation of the patient be-
gins with a history of urinary complaints, the pa-
tient's description of symptoms may be mislead-
ing. In a study by Glezerman et al. (55), stress
incontinence was diagnosed in 81.6% of 130
women complaining of urinary stress inconti-
nence, and it was concluded that history alone
would not have been an appropriate indication for
surgery. The history should not be solely relied on
in determining therapy for incontinence. Urody-
namic evaluation may reveal unsuspected blad-
der dysfunction. An assumption cannot be made
that all women have stress incontinence, espe-
cially those patients with neurologic disease.

The significance of the presence of detrusor
instability in the setting of stress incontinence
needs to be understood when considering surgical
therapy. Stress incontinence and detrusor insta-
bility should be differentiated in the evaluation of
urinary incontinence. In a study of 309 elderly
incontinent patients, pure stress incontinence
was rarely found (22). Detrusor instability is of-
ten the cause of leakage of urine associated with
coughing or laughing (22). In some incontinent
women, both stress incontinence and detrusor in-
stability may be present (53). Drutz and Mandel
(56) reported that, of women with a history sug-
gestive of pure sphincter weakness incontinence,
over 20% had unstable bladders. Both stress in-
continence and urgency incontinence may be
cured by surgical elevation of the vesical neck
(53). Although there are reports of postoperative
failures in women with detrusor instability (57),
the presence of detrusor instability in a patient
with stress incontinence is not predictive of a poor
operative result (31, 53); on the other hand, pa-
tients with detrusor instability who do not have

anatomic stress incontinence may not benefit
from surgery. A small number of patients without
stress incontinence may be detected by urody-
namic and radiographic studies (31). Many older
women may be physically ineligible for surgery or
fearful of operations (58).

Any of the symptoms of detrusor instability
may be confused with anatomic stress inconti-
nence, since an increase in intra-abdominal pres-
sure may precipitate an uninhibited detrusor con-
traction. Since treatment plans may differ, it is
important to differentiate between them. Al-
though urine loss associated with coughing may
be visually demonstrated in patients with stress
incontinence, a spontaneous detrusor contraction
can occur 5-15 sec after a cough (53). Since de-
trusor instability may not be detected in the su-
pine position, it is necessary to perform cystome-
try with postural change and with coughing (57).
Because it is difficult to classify the etiology of
urinary incontinence based on symptoms alone,
cystometric studies have been recommended in
the evaluation of patients prior to surgical treat-
ment for stress incontinence (23, 59). This is es-
pecially important in the evaluation of inconti-
nent women with neurologic disorders, who may
have detrusor instability alone or in addition to
anatomic stress incontinence (60).

The simple provocative full-bladder stress
test is as effective as the radiographic or elec-
tronic pressure measurement in detecting ure-
thral incompetence producing stress urinary in-
continence (25), but frail elderly patients may not
be able to cough, strain, or assume the standing
position (61). If provocative full-bladder testing is
negative in eliciting incontinence, the other tests
should be done (25). The Q-tip test is nonspecific,
since many postmenopausal patients without
stress incontinence have a positive test (59).

Functional Incontinence. Urinary inconti-
nence can occur in the presence of normal bladder
function if the patient is unable to be toileted in
an appropriate and timely manner. There is a re-
lationship between urinary incontinence and fac-
tors such as poor mobility and confusion (62).

Urinary incontinence among elderly patients
in the hospital setting is associated with impaired
cognitive and physical functioning (63-65). In
nursing home patients, urinary incontinence has
been found to be more closely associated with im-
pairments of physical and mental function than
with primary bladder problems (17, 66). Since
this strong relationship between mobility prob-
lems and urinary incontinence exists in elderly
patients (67), it is not surprising to find that func-

Vol. 60 No. 6

URINARY INCONTINENCE IN THE ELDERLY— ST ARER

505

tional incontinence, secondary to immobility, is a
major problem in the nursing home population

(68).

Urinary incontinence has been described as
"the result of a predisposing factor and a precip-
itating factor," in which the predisposing factor
could be detrusor instability (12). If patients are
able to alter their environment and lifestyle to
meet the demands of an unstable bladder, they
may remain continent (12, 69). The loss of mobil-
ity or mental clarity, in addition to the presence
of detrusor instability, can result in urinary in-
continence (70). Detrusor instability in an elderly
patient by itself may not be sufficient to cause
incontinence (32, 33), but with additional stress
such as dementia or immobility, incontinence
may occur (14). The use of anticholinergic medi-
cations may not alleviate incontinence in institu-
tionalized patients unless there is also a toileting
program to compensate for a patient's limited mo-
bility or cognitive dysfunction (17).

Entities that Defy Classification

Postprostatectomy Incontinence. Conti-
nence after radical retropubic prostatectomy is
dependent on sphincteric efficiency (71). Preser-
vation of continence following radical retropubic
prostatectomy requires a functional urethral
length of at least 2.8 cm and a closed, nonob-
structed bladder neck (72). Factors such as short-
ened urethral functional length, decreased ure-
thral closure pressure, and urethral mobility are
important in postradical perineal prostatectomy
incontinence (73). Approximately 38% of patients
report incontinence during the first three months
following transurethral resection of the prostate
(35). Although in the parkinsonian patient the
major risk of incontinence following transure-
thral prostatectomy is associated with lack of vol-
untary sphincter control (74), the cause of incon-
tinence after prostatectomy (radical, suprapubic,
or transurethral resection) may be bladder dys-
function (detrusor instability or low bladder wall
compliance) (75, 76).

The preoperative finding of detrusor instabil-
ity should alert the physician to the possibility of
prolonged postoperative symptoms (28). However,
the detrusor response eventually will become nor-
mal, and symptoms will disappear in almost all of
these patients (28). Although patients who com-
plain of urgency, frequency, and nocturia before
prostatectomy are more likely to have detrusor
instability, the postprostatectomy symptoms do
not correlate well with urodynamic findings (77).

Persistent postoperative involuntary detrusor
contractions are associated with an unfavorable
surgical outcome (78).

Comprehensive urodynamic evaluation is es-
sential before treatment can be recommended for
incontinence after prostatectomy (75, 76). Urine
flow rate measurement may be used to isolate a
group of patients with prostatism symptoms but
no bladder outflow obstruction who are unlikely
to benefit from surgery (79). A normal flow rate
suggests the possibility of nonobstructive bladder
disease, such as detrusor instability, whereas a
reduced urine flow rate is evidence of bladder out-
flow obstruction (79).

Although preoperative cystometric findings
cannot predict which patients will have involun-
tary detrusor contractions following prostatec-
tomy (78), it has been recommended that all pa-
tients with dementia or a major neurologic
disorder be evaluated by urodynamic testing be-
fore transurethral prostatectomy (80). Others
have suggested that patients with bladder out-
flow obstruction with a high urgency symptom
score (frequency, nocturia, enuresis, urgency) and
an unexpectedly high flow rate (greater than 12
mL/sec) need preoperative urodynamic studies
(29). Fitzpatrick et al. (77) recommend a urody-
namic assessment before prostatectomy for all pa-
tients with a combination of frequency, nocturia,
and urgency. Kimche et al. (81) recommend that
patients with benign prostatic hypertrophy and
detrusor instability undergo an examination of
the evoked response of the third neural loop in the
genitourinary system to ascertain whether there
is deterioration of neural function that could
cause the instability. Those patients with a low
evoked response should not have a prostatectomy
(81).

Cerebrovascular Accident. Bladder dysfunc-
tion may occur following a cerebrovascular acci-
dent (CVA) (82). Although detrusor hyperre-
flexia, the most common cystometric abnormality
in patients with voiding problems following a
CVA (83, 84), has been explained by the loss of
voluntary control of the reflex arc between the
bladder and the sacral spinal cord (14), urinary
retention also occurs in patients who have had a
CVA. In a study by Borrie et al. (85), 22 post-
stroke patients with urinary incontinence under-
went cystometric studies. The majority had detru-
sor hyperreflexia, but two patients were reported
as being in urinary retention.

Damage to the frontal lobe or internal cap-
sule results in detrusor hyperreflexia (86), which
has also been associated with subclinical lesions

506

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

of the basal ganglia on magnetic resonance imag-
ing (87). The finding of both inhibitory and facil-
itatory influences by the brain on the sacral
micturition reflex arc suggests the complexity in-
volved in predicting bladder dysfunction that
may occur in the setting of destructive or irrita-
tive neurologic lesions (88). In stroke patients,
the contribution to urinary incontinence of cogni-
tive impairment and immobility as well as of neu-
rologic injury needs to be considered (89).

Parkinson's Disease. In Parkinson's disease
there is a high incidence of bladder dysfunction
(90). Detrusor hyperreflexia has been reported to
be the most common urodynamic pattern in pa-
tients with Parkinson's disease (91-97). Volun-
tary control of voiding occurs via the influence of
the cortical micturition center on the pontine-
mesencephalic reticular formation. The thalamic
nuclei, the basal ganglia, and the anterior vermis
of the cerebellum also have input into this system
(94). In Parkinson's disease, the neurologic lesion
in the basal ganglia interferes with voluntary
control of the micturition reflex (92). Anticholin-
ergic medications have been reported as useful in
the treatment of urinary-incontinent patients
with Parkinson's disease (98).

Diabetes Mellitus. In diabetic patients, uri-
nary retention secondary to autonomic neuropa-
thy may be the cause of incontinence (14,
99-101). The loss of bladder sensation, which has
been described in diabetics (102-104), results
from damage to the afferent nerves from the blad-
der to the central nervous system (105). This loss
of sensation leads to distension of the bladder,
decompensation of the detrusor muscle, and uri-
nary retention (106, 107). Treatment is directed
at insufficient bladder emptying (108).

Both poor bladder contractility and involun-
tary bladder contractions have been observed in
diabetic patients (109-114). Bradley (115) de-
scribes two bladder abnormalities in diabetic pa-
tients: detrusor areflexia with impaired sensa-
tion, and involuntary contractions of the bladder.
The key question is which disorder is responsible
for the patient's symptoms. In a diabetic patient
with urinary symptoms it seems unlikely that the
cause can be accurately determined without fur-
ther investigation, such as urodynamic studies
(114).

Scheduled voiding may be effective for the
diabetic patient with a sensory abnormality that
results in infrequent voiding; clean intermittent
catheterization may be used for the diabetic pa-
tient with decreased bladder contractility (116).
Careful thought must be given to the decision to
suppress involuntary contractions with anti-

cholinergic medications, especially in the diabetic
patient who may be prone to urinary retention.
Similarly, any medication with anticholinergic
activity, such as tricyclic antidepressants, should
be used cautiously in diabetic patients (117). The
situation may be complicated by a recently de-
scribed entity in which the bladder contracts in-
voluntarily but is incompletely emptied (118), a
condition that also may predispose to anticholin-
ergic-induced retention. Urodynamic studies can
be helpful when choosing therapy for the elderly
diabetic.

Evaluation of Incontinence

Residual Urine. The residual urine is the
urine remaining in the bladder following urina-
tion. Measurement of the postvoiding residual
urine is a simple way to assess "the effectiveness
of bladder emptying" (119). An elevated postvoid-
ing residual urine may indicate bladder outlet ob-
struction or ineffective detrusor contractions (14).
In discussions about urinary incontinence in the
elderly, values for a significant volume of resid-
ual urine have ranged from 20 mL to 300 mL (13,
14, 120, 121).

It may be difficult to interpret postvoiding
residual urine measurements in the evaluation of
urinary incontinence, since the normal residual
urine in elderly individuals is uncertain. Studies
in incontinent women over age 60 have shown
that 13% to 29% have residual urine greater than
100 mL (24, 120), and a study of continent women
over age 60 reported 13% having residual urines
greater than 100 mL (32). In a study of continent
men over age 60 (33), only one subject was
reported as having residual urine greater than
100 mL.

It is not always possible to predict the type of
bladder dysfunction based solely on a measure-
ment of residual urine. In a study of 49 adult men
with symptoms of prostatism, Bruskewitz et al.
(122) found no correlation between preoperative
residual urine volume and the urodynamic find-
ings. Bladder outlet obstruction does not need to
be present for a patient to have elevated residual
urine. An incomplete bladder contraction that
terminates before emptying is complete can ele-
vate postvoiding residual urine (123). Uninhib-
ited contractions may not efficiently empty the
bladder (118). Therefore, even without outflow
obstruction, urinary retention may occur. On the
other hand, there may be unsuspected causes of
bladder outlet obstruction. When the periurethral
striated muscle relaxes during an uninhibited de-
trusor contraction, complete emptying can occur.

Vol. 60 No. 6

URINARY INCONTINENCE IN THE ELDERLY— STARER

507

However, if that muscle contracts during unin-
hibited contraction, the volume of residual urine
will be elevated (124).

There can be problems in obtaining an accu-
rate measurement of the postvoiding residual
volume of urine in many elderly patients (125).
For one thing, these individuals may not be able
to void on command in a testing situation, leading
to uncertainty over whether the inability to void
spontaneously reflects bladder dysfunction or
nervousness. The interpretation of the catheter-
ized urine volume must take this into account.
For another, Bruskewitz et al. (122) have noted
that measured volumes varied widely in the same
individual when postvoiding residual urine deter-
minations were repeated. Although it is sug-
gested that a single measurement may not be de-
finitive, some patients may be reluctant to be
catheterized several times.

The measurement of residual urine is an im-
portant tool in the evaluation of bladder function.
Residual urine volumes less than 100 mL usually
indicate normal bladder emptying. High volumes
(300 mL or more) indicate problems with bladder
emptying. If detrusor instability is found in addi-
tion to elevated residual urine, an obstruction of
the bladder outlet should be considered. Bladder
outflow obstruction can occur in women (126).
The residual urine measurement is useful for de-
termining which patients should be referred for
further testing.

An accurate residual urine measurement re-
quires that the patient have an adequate oppor-
tunity to void prior to catheterization, but be-
cause a true measurement of the postvoiding
residual urine in an elderly patient may be diffi-
cult, the residual urine measurement is most use-
ful when considered along with all other avail-
able data.

Urodynamic Studies. Urodynamic studies
can be useful in the evaluation and treatment of
incontinence in the elderly (21, 127), but they are
not without cost and risk (128), and some contro-
versy exists concerning using urodynamic studies
to evaluate incontinent elderly patients (129). A
technologically simpler approach to evaluation
may be desirable (120). If the cause of inconti-
nence could be determined by history and nonin-
vasive examination, then urodynamic studies
could possibly be avoided (100). Simple cystome-
try can be helpful to assess bladder function
among incontinent geriatric patients in settings
in which formal urodynamics are either unavail-
able or impractical (130).

Since there is a poor correlation between the
symptoms of urinary incontinence and urody-

namic abnormalities, patients may require fur-
ther investigation to determine the cause of the
problem and to decide on the treatment (21, 131).
A medical history and physical examination are
helpful in predicting urethral incompetence in
patients with simple, clear-cut symptoms of exer-
tional or stress incontinence (25). When the
symptoms are urge, mixed stress and urge, or
suggestive of overflow incontinence, cystometry,
residual urine measurement, and provocative
full-bladder stress testing are needed (25). Single-
channel cystometry, when used in conjunction
with simple urethral tests, can provide most of
the information that is obtained with multichan-
nel cystometry, in which abdominal pressure is
electronically subtracted from total bladder pres-
sure (132), and visual confirmation of urine leak-
age with coughing is useful (132). Complex uro-
dynamic tests should be reserved for complex
symptomatology (25).

Some cases of detrusor instability diagnosed
by cystometry may be false positives. Detrusor
instability can be provoked by the use of a gas at
a rapid filling rate (34), and the clinical relevance
of this requires further investigation. Day-to-day
variability in cystometric findings has been re-
ported in elderly patients (61).

Cystometric improvements are not synony-
mous with clinical benefit (133).

Cystometry: Bladder Sensation, The mea-
surement of volume at first sensation that is ob-
tained during cystometry can be difficult to inter-
pret unless it is markedly elevated. The "normal"
volume for sensing bladder filling is not clear. In
a group of normal adults ranging in age from 16
to 85 years, Merrill et al. (134) described the vol-
ume at first sensation as ranging widely from 30
mL to 460 mL (mean, 137 mL). Andersen et al.
(33) studied a group of healthy elderly men and
found the bladder volume at first sensation to
range from 20 mL to 200 mL (mean, 127 mL). It
has been pointed out that because sensation is
subjective, it is not easy to evaluate (135). The
patient's ability to sense bladder filling is difficult
to assess, although it is an important component
of bladder function.

Urinary Flow Rate. In men, the peak urinary
flow rate decreases with increasing age (136), but
for normal women, it does not deteriorate much
(137); the decrease in men is an average of 2 mL/
sec per decade, whereas in women it is only 0.3
mL/sec per decade (137). Although most men with
bladder outlet obstruction have a diminished flow
rate, uroflowmetry as a single examination can-
not distinguish between bladder outlet obstruc-
tion and impaired detrusor contractility (138).

508 THE MOUNT SINAI JOURNAL OF MEDICINE November 1993

Urethral Pressure Profile. The role of the
urethral pressure profile in the evaluation of el-
derly incontinent patients is uncertain (15).

Management

Diapers. Although treatment can be directed
at the specific cause in an attempt to eradicate
incontinence, in recent years the use of absorptive
pads has increased (12, 139). The disposable dia-
per, which has become popular in the care of the
incontinent patient, has advantages and disad-
vantages.

The adult diaper is used frequently for incon-
tinent nursing home patients (10). Although the
diaper simplifies the nursing care of the inconti-
nent patient (140, 141) and allows increased mo-
bility and socialization by the patient (142), it
does not cure incontinence (143). The diaper may
be difficult to remove for the elderly patient who
wishes to void at the toilet. In addition, the diaper
may indicate to both the staff and the patient that
incontinence is medically acceptable. Some indi-
viduals experience a loss of dignity when wearing
a diaper because it is associated with infancy (12,
142). The cost of using a disposable product for
the long-term must also be considered (142, 144).

Toileting requires more staff time than does
the changing of a diaper (145), and skin irritation
and breakdown may develop if diapers are not
changed at appropriate intervals (12). Before di-
apers are used, an attempt should be made to de-
termine the cause of incontinence and to use that
information to choose an effective treatment. Per-
manent use of diapers should be a treatment of
last resort (14). In cases where specific treatment
fails, diapers may be the only alternative (12).

Stress Incontinence. Surgery is an effective
treatment of pure stress incontinence associated
with urethrocele (146). The goal of most opera-
tions is to elevate the bladder neck to within the
abdominal zone of pressure (147). Urethral sling
procedures pass a ribbon of fascia or artificial ma-
terial beneath the urethra, which elevates and
compresses it (146). Although norephedrine, an
alpha-adrenoreceptor stimulating agent, pro-
duces a significant increase in maximum urethral
closure pressure, the therapeutic effect is moder-
ate (148). Oral estriol and phenylpropanolamine
in combination have been recommended for the
treatment of stress urinary incontinence in post-
menopausal women (149).

Pelvic Floor Muscle Exercise. Pelvic muscle
exercise (perivaginal muscle contraction and re-
laxation) is advocated to prevent and treat stress
urinary incontinence, for it is thought to increase

urethral resistance by increasing periurethral
muscle tension (150). It has been recommended
that exercises (a 10-sec pubococcygeal contraction
followed by a 10-sec relaxation) be done at least
80 times per day (58). This therapy may not be
applicable to most elderly patients.

Behavioral Treatment. Behavioral treat-
ment requires that patients be able to learn new
skills and participate actively in their own treat-
ment for several weeks (151). Behavioral training
with biofeedback should be considered among the
first treatments offered to nondemented patients
with stress or urge incontinence (152, 153). Bio-
feedback enables suitably motivated patients to
learn voluntary control by observing the results
of attempts to control bladder, sphincter, and ab-
dominal pressure responses (152). For stress in-
continence, the primary goal of training is to en-
able the patient to contract the periurethral
muscles selectively while inhibiting contraction
of abdominal muscles; for urge incontinence, the
goals are to train the patient to inhibit detrusor
contraction voluntarily and to contract periure-
thral muscles selectively to prevent urine loss un-
til detrusor inhibition can be achieved (152).

Scheduling regimens used as behavioral in-
terventions for urinary incontinence include
bladder training, habit retraining, timed voiding,
and prompted voiding (154). In bladder training,
the patient progressively extends the intervoid-
ing interval; in habit retraining, the toileting
schedule is adjusted to fit the patient's individual
voiding pattern (episodes of incontinence, pa-
tient's desires to void); in timed voiding, the pa-
tient is given a fixed voiding schedule that re-
mains unchanged; with prompted voiding, the
patient is asked at regular intervals if she needs
to void and is assisted in using the toilet only if
the response is positive (154).

Urinary incontinence in elderly, nonde-
mented, ambulatory, community-dwelling pa-
tients can be effectively treated by a nurse prac-
titioner and internist/geriatrician utilizing
behavioral training with or without bladder-
sphincter biofeedback (155). Bladder training is
beneficial in noninstitutionalized older women
with stress incontinence alone, those with detru-
sor instability alone, and those with both detrusor
instability and stress incontinence (156). How-
ever, bladder retraining programs in the nursing
home setting may require an impractical amount
of nursing care (157).

Detrusor Instability. In some cases, involun-
tary contractions of the bladder may be treated
with bladder retraining or anticholinergic medi-
cations. Urinary tract infection should be treated

Vol. 60 No. 6

URINARY INCONTINENCE IN THE ELDERLY— ST ARER

509

with antibiotics. Patients with involuntary blad-
der contractions may require additional urologic
evaluation to determine the cause of the dysfunc-
tion— inflammation, neoplasm, prostatic hyper-
trophy.

Drug therapy plays a minor role in the treat-
ment of detrusor instability (158). Anticholin-
ergic medications have been utilized in the treat-
ment of detrusor instability. Oxybutynin chloride
has both an anticholinergic effect and a direct
spasmolytic effect on the bladder smooth muscle
(159). Oxybutynin chloride at dosages of 2.5-5
mg, three times per day, is safe for use in the
elderly; dryness of the mouth is the most common
side effect (160). There have been a limited num-
ber of clinical trials of oxybutynin in incontinent
elderly patients. Neither a habit-training pro-
gram alone nor one combined with oxybutynin
was shown to have any significant effects on the
frequency of incontinence in a placebo-controlled
trial involving functionally impaired nursing
home patients with detrusor hyperreflexia (161).
In a study of elderly institutionalized patients
with detrusor instability, oxybutynin was no
more effective than placebo for the treatment of
incontinence (162). In an uncontrolled study of
the effects of oxybutynin in eight adult patients
with involuntary contractions of the detrusor, Di-
okno and Lapides (163) demonstrated symptom-
atic improvement in all patients and urodynamic
improvement in seven of the eight. Moisey et al.
(159) conducted a controlled study of 23 patients
with detrusor instability (age range, 20-79 years)
in which patients received 28-day courses of pla-
cebo or oxybutynin chloride (5 mg three times a
day). Although 17 patients (69%) had symptom-
atic improvement, urodynamic changes did not
match the symptomatic improvement in most of
them. Oxybutynin chloride administered intra-
vesically has been suggested for the management
of patients who do not respond to oral mediations
or who have intolerable side effects (164).

Although imipramine has a place in the
treatment of urinary incontinence associated
with detrusor instability in the elderly (133), el-
derly incontinent patients do not do significantly
better with imipramine and bladder training
than with bladder training alone (158). Elderly
patients with urinary incontinence and detrusor
instability who are treated with bladder training
and imipramine do well if they have good cogni-
tive function and good mobility (165). Patients
are easier to cure if detrusor instability is less
severe — that is, the involuntary contraction oc-
curs at a higher volume, reaches a lower pressure,
or is sustained for a shorter period (165).

Pinacidil, which inhibits contractile activity
in smooth muscle by the opening of potassium
channels, has not been effective for the treatment
of unstable detrusor contractions associated with
bladder outflow obstruction (166).

The aim of regular toileting is to prevent in-
continence by emptying the bladder before the
contents reach the volume at which an involun-
tary contraction occurs (167). Toileting may fail
in patients with small bladder capacities (invol-
untary contractions occurring at volumes less
than 150 mL) because toileting more frequently
than every two hours is difficult (167). Even if
incontinence occurs, toileting must be done regu-
larly because some urine may remain in the blad-
der if the involuntary contraction is not sustained
(167).

Nursing Homes. Nonurological problems (be-
havioral problems, cognitive dysfunction, immo-
bility, medication problems, diabetes, local pa-
thology), which frequently contribute to urinary
incontinence in long-term care facilities (17, 61,
66, 168), should be addressed prior to considering
urodynamic studies (168). The frequency of incon-
tinence was reduced in nursing homes with a be-
havioral approach whereby patients were
"prompted" hourly and toileted only when they
asked for assistance (169-171). Some of the im-
provement may have occurred because the pa-
tients learned to request help (171). Patients with
normal bladder capacity and patients who were
more cognitively intact responded better to the
behavior therapy program (171). Electrical stim-
ulation with a rectal probe is not effective by it-
self in improving urinary incontinence in func-
tionally impaired nursing home patients (157).

Unfortunately, a regular toileting schedule
may be difficult for the nursing staff to maintain
(145), and toileting programs are more likely to
be successful with either an increase in trained
personnel or a more productive redistribution of
the work (172). The systematic toileting of pa-
tients may improve the quality of patients' lives,
although the nursing home may not experience a
savings in labor and supply cost (173).

Summary

The training of health care providers in the
diagnosis and management of urinary inconti-
nence is inadequate, and evaluation and manage-
ment skills are less than optimal (22). Urinary
incontinence volume measurement is difficult to
perform accurately in elderly long-term-care in-
patients (174). After incontinence has been noted,
the physician needs to select the appropriate

510

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

studies to determine the cause and choose the cor-
rect treatment. Otherwise, treatable causes such
as outlet obstruction or fecal impaction may be
missed. Subsequent management includes sched-
uled visits to the toilet, treatment of impaired mo-
bility, use of easily identified and accessible toilet
facilities, reevaluation of the need for restraints
and side rails, and, if appropriate, the use of drugs
to treat the incontinence. Many elderly inconti-
nent patients who are treated in the outpatient
setting can be cured or substantially improved
(15).

References

1. Yarnell JWG, St. Leger AS. The prevalence, severity and

factors associated with urinary incontinence in a ran-
dom sample of the elderly. Age Ageing 1979; 8:81-85.

2. Thomas TM, Plymat KR, Blannin J, Meade TW. Preva-

lence of urinary incontinence. Br Med J 1980; 281:
1243-1245.

3. Feneley RCL, Shepherd AM, Powell PH, Blannin J. Uri-

nary incontinence: prevalence and needs. Br J Urol
1979; 51:493-496.

4. Vetter NJ, Jones DA, Victor CR. Urinary incontinence

in the elderly at home. Lancet 1981; 2:1275-1277.

5. Yarnell JWG, Voyle GJ, Richards CJ, Stephenson TP.

The prevalence and severity of urinary incontinence in
women. J Epidemiol Community Health 1981; 35:71-
74.

6. Herzog AR, Diokno AC, Brown MB, Normolle DP, Brock

BM. Two-year incidence, remission, and change pat-
terns of urinary incontinence in noninstitutionalized
older adults. J Gerontol Med Sci 1990; 45(2):M67-74.

7. Ouslander JG, Kane RL, Abrass IB. Urinary inconti-

nence in elderly nursing home patients. JAMA 1982;
248:1194-1198.

8. Sogbein SK, Awad SA. Behavioral treatment of urinary

incontinence in geriatric patients. Can Med Assoc J
1982; 127:863-864.

9. Marron KR, Fillit H, Peskowitz M, Silverstone FA. The

non-use of urethral catheterization in the manage-
ment of urinary incontinence in the teaching nursing
home. J Am Geriatr Soc 1983; 31:278-281.

10. Starer P, Libow LS. Obscuring urinary incontinence.

Diapering of the elderly. J Am Geriatr Soc 1985; 33:
842-846.

11. Williams ME. A critical evaluation of the assessment

technology for urinary continence in older persons. J
Am Geriatr Soc 1983; 31:657-664.

12. Brocklehurst JC. The urinary tract. In: Rossman I, ed.

Clinical geriatrics. Philadelphia: JB Lippincott, 1979:
317-330.

13. Ouslander JG. Urinary incontinence in the elderly. West

J Med 1981; 135:482-496.

14. Williams ME, Pannill FC. Urinary incontinence in the

elderly. Ann Intern Med 1982; 97:895-907.

15. Ouslander JG, Hepps K, Raz S, Su H-L. Genitourinary

dysfunction in a geriatric outpatient population. J Am
Geriatr Soc 1986; 34:507-514.

16. Resnick NM, Yalla SV, Laurino E. The pathophysiology

of urinary incontinence among institutionalized elder-
ly persons. N Engl J Med 1989; 320:1-7.

17. Ouslander JG, Uman GC, Urman HN, Rubenstein LZ.

Incontinence among nursing home patients: clinical

and functional correlates. J Am Geriatr Soc 1987; 35:
324-330.

18. Hebjorn S, Andersen JT, Walter S, Dam AM. Detrusor

hyperreflexia. A survey on its etiology and treatment.
Scand J Urol Nephrol 1976; 10:103-109.

19. Tang PC. Levels of brain stem and diencephalon control-

ling micturition reflex. J Neurophysiol 1955; 18:583-
595.

20. Tang PC, Ruch TC. Localization of brain stem and dien-

cephalic areas controlling the micturition reflex. J
Comparative Neurol 1956; 106:213-231.

21. Castleden CM, Duffin HM, Asher MJ. Clinical and uro-

dynamic studies in 100 elderly incontinent patients.
Br Med J 1981; 282:1103-1105.

22. Overstall PW, Rounce K, Palmer JH. Experience with an

incontinence clinic. J Am Geriatr Soc 1980; 28:535-
538.

23. Eastwood HDH, Warrell R. Urinary incontinence in the

elderly female: prediction in diagnosis and outcome of
management. Age Ageing 1984; 13:230-234.

24. Brocklehurst JC, Dillane JB. Studies of the female blad-

der in old age. II. Cystometrograms in 100 incontinent
women. Gerontol Clin 1966; 8:306-319.

25. Diokno AC, Wells TJ, Brink CA. Urinary incontinence

in elderly women: urodynamic evaluation. J Am Geri-
atr Soc 1987; 35:940-946.

26. Turner-Warwick R, Whiteside CG, Arnold EP, et al. A

urodynamic view of prostatic obstruction and the re-
sults of prostatectomy. Br J Urol 1973; 45:631-645.

27. Andersen JT. Detrusor hyperreflexia in benign intraves-

ical obstruction. A cystometric study. J Urol 1976; 115:
532-534.

28. Cote RJ, Burke H, Schoenberg HW. Prediction of un-

usual postoperative results by urodynamic testing in
benign prostatic hyperplasia. J Urol 1981; 125:690-
692.

29. Speakman MJ, Sethia KK, Fellows GJ, Smith JC. A

study of the pathogenesis, urodynamic assessment and
outcome of detrusor instability associated with blad-
der outflow obstruction. Br J Urol 1987; 59:40^4.

30. Andersen JT. Prostatism. III. Detrusor hyperreflexia

and residual urine. Clinical and urodynamic aspects
and the influence of surgery on the prostate. Scand J
Urol Nephrol 1982; 16:25-30.

31. McGuire EJ, Lytton B, Kohorn EI, Pepe V. The value of

urodynamic testing in stress urinary incontinence. J
Urol 1980; 124:256-258.

32. Brocklehurst JC, Dillane JB. Studies of the female blad-

der in old age. I. Cystometrograms in non-incontinent
women. Gerontol Clin (Basel) 1966; 8:285-305.

33. Andersen JT, Jacobsen O, Worm-Petersen J, Hald T.

Bladder function in healthy elderly males. Scand J
Urol Nephrol 1978; 12:123-127.

34. McGuire EJ. Relevant clinical urodynamics. In: McGuire

EJ, ed. Urinary incontinence. New York: Grune &
Stratton, 1981:37-51.

35. Doll HA, Black NA, McPherson K, Flood AB, Williams

GB, Smith JC. Mortality, morbidity and complications
following transurethral resection of the prostate for
benign prostatic hypertrophy. J Urol 1992; 147:1566-
1573.

36. Abrams PH, Farrar DJ, Turner-Warwick RT, Whiteside

CG, Feneley RCL. The results of prostatectomy: a
symptomatic and urodynamic analysis of 152 patients.
J Urol 1979; 121:640-642.

37. Andersen JT, Nordling J, Walter S. Prostatism. I. The

correlation between symptoms, cystometric and uro-

Vol. 60 No. 6 URINARY INCONTINENCE IN THE ELDERLY— STARER 511

dynamic findings. Scand J Urol Nephrol 1979; 13:229-
236.

38. Lum SK, Marshall VR. Results of prostatectomy in pa-

tients following a cerebrovascular accident. Br J Urol
1982; 54:186-189,

39. Moisey CU, Rees RWM. Results of transurethral resec-

tion of prostate in patients with cerebrovascular dis-
ease. Br J Urol 1978; 50:539-541.

40. Finkbeiner AE. Is bethanechol chloride clinically effec-

tive in promoting bladder emptying? A literature re-
view. J Urol 1985; 134:443-449.

41. Barbalias GA, Klauber GT, Blaivas JG. Critical evalua-

tion of the crede maneuver: a urodynamic study of 207
patients. J Urol 1983; 130:720-723.

42. Lepor H. Nonoperative management of benign prostatic

hyperplasia. J Urol 1989; 141:1283-1289.

43. Caine M, Perlberg S, Meretyk S. A placebo-controlled

double-blind study of the effect of phenoxybenzamine
in benign prostatic obstruction. Br J Urol 1978; 50:
551-554.

44. Abrams PH, Shah PJR, Stone R, Choa RG. Bladder out-

flow obstruction treated with phenoxybenzamine. Br J
Urol 1982; 54:527-530.

45. Lepor H, Meretyk S, Knapp-Maloney G. The safety, ef-

ficacy and compliance of terazosin therapy for benign
prostatic hyperplasia. J Urol 1992; 147:1554-1557.

46. Gormley GJ, Stoner E, Bruskewitz RC, et al. The effect of

finasteride in men with benign prostatic hyperplasia.
N Engl J Med 1992; 327:1185-1191.

47. Yalla SV, Blunt KJ, Fam BA, Constantinople NL, Gittes

RF. Detrusor-urethral sphincter dyssynergia. J Urol
1977; 118:1026-1029.

48. Blaivas JG, Sinha HP, Zayed AAH, Labib KB. Detrusor-

external sphincter dyssynergia. J Urol 1981; 125:542-
544.

49. Leyson JFJ, Martin BF, Sporer A. Baclofen in the treat-

ment of detrusor-sphincter dyssynergia in spinal cord
injury patients. J Urol 1980; 124:82-84.

50. McGuire EJ, Brady S. Detrusor-sphincter dyssynergia. J

Urol 1979; 121:774-777.

51. Siroky MB, Krane RJ. Neurologic aspects of detrusor-

sphincter dyssynergia, with reference to the guarding
refiex. J Urol 1982; 127:953-957.

52. Rudy DC, Awad SA, Downie JW. External sphincter dys-

synergia: an abnormal continence reflex. J Urol 1988;
140:105-110.

53. Shortliffe LMD, Stamey TA. Urinary incontinence in the

female. Stress urinary incontinence. In: Walsh PC,
Gittes RF, Perlmutter AD, Stamey TA, eds. Camp-
bell's urology. Philadelphia: WB Saunders, 1986:
2680-2711.

54. Green TH. Urinary stress incontinence: differential di-

agnosis, pathophysiology, and management. Am J Ob-
stet Gynecol 1975; 122:368-400.

55. Glezerman M, Glasner M, Rikover M, Tauber E, Bar-Ziv

J, Insler V. Evaluation of reliability of history in
women complaining of urinary stress incontinence.
Eur J Obstet Gynecol Reprod Biol 1986; 21:159-164.

56. Drutz HP, Mandel F. Urodynamic analysis of urinary

incontinence symptoms in women. Am J Obstet Gyne-
col 1979; 134:789-792.

57. Arnold EP, Webster JR, Loose H, Brown ADG, Turner-

Warwick RT, Whiteside CG, Jequier AM. Urodynam-
ics of female incontinence: factors influencing the re-
sults of surgery. Am J Obstet Gynecol 1973; 117:805-
813.

58. Wells TJ, Brink CA, Diokno AC, Wolfe R, Gillis GL.

Pelvic muscle exercise for stress urinary incontinence
in elderly women. J Am Geriatr Soc 1991; 39:78.5-791.

59. Bent AE, Richardson DA, Ostergard DR. Diagnosis of

lower urinary tract disorders in postmenopausal pa-
tients. Am J Obstet Gynecol 1983; 145:218-222,

60. Khan Z, Starer P, Bhola A. Urinary incontinence in fe-

male Parkinson's disea.se patients. Pitfalls of diagno-
sis. Urology 1989; 33:486-^89.

61. Dennis PJ, Igou JF, Rohner Jr. TJ, Yu LC, Hu T-W,

Kaltreider DL. Simple urodynamic evaluation of in-
continent elderly female nursing home patients. A de-
scriptive analysis. Urology 1991; 37:173-179.

62. Campbell AJ, Reinken J, McCosh L. Incontinence in the

elderly: prevalence and prognosis. Age Aging 1985;
14:65-70.

63. Sier H, Ouslander J, Orzeck S. Urinary incontinence

among geriatric patients in an acute-care hospital.
JAMA 1987; 257:1767-1771.

64. Ekelund P, Rundgren A. Urinary incontinence in the

elderly with implications for hospital care consump-
tion and social disability. Arch Gerontol Geriatr 1987;
6:11-18.

65. Fernie GR, Jewett MAS, Autry D, HoUiday PG, Zorzitto

ML. Prevalence of geriatric urinary dysfunction in a
chronic care hospital. Can Med Assoc J 1983; 128:
1085-1086.

66. Yu LC, Rohner TJ, Kaltreider DL, Hu T-W, Igou JF,

Dennis PJ. Profile of urinary incontinent elderly in
long-term care institutions. J Am Geriatr Soc 1990;
38:433^39.

67. Diokno AC, Brock BM, Herzog AR, Bromberg J. Medical

correlates of urinary incontinence in the elderly. Urol-
ogy 1990; 36:129-138.

68. Ribeiro BJ, Smith SR. Evaluation of urinary catheter-

ization and urinary incontinence in a general nursing
home population. J Am Geriatr Soc 1985; 33:479-482.

69. Diokno AC, Brown MB, Brock BM, et al. Clinical and

cystometric characteristics of continent and inconti-
nent noninstitutionalized elderly. J Urol 1988; 140:
567-571.

70. Rowe JW, Resnick NM. Disorders of the kidney and uri-

nary tract. In: Andres R, Bierman EL, Hazzard WR,
eds. Principles of geriatric medicine. New York: Mc-
Graw-Hill, 1985:614-628.

71. Presti Jr. JC, Schmidt RA, Narayan PA, Carroll PR,

Tanagho EA. Pathophysiology of urinary incontinence
after radical prostatectomy. J Urol 1990; 143:975-978.

72. Rudy DC, Woodside JR, Crawford ED. Urodynamic eval-

uation of incontinence in patients undergoing modi-
fied Campbell radical retropubic prostatectomy: a pro-
spective study. J Urol 1984; 132:708-712.

73. Krauss DJ, Paletsky SH, Lilien OM. Urodynamics of

post-radical perineal prostatectomy incontinence. J
Urol 1980; 124:263-265.

74. Staskin DS, Vardi Y, Siroky MB. Post-prostatectomy

continence in the parkinsonian patient: the signifi-
cance of poor voluntary sphincter control. J Urol 1988;
140:117-118.

75. Leach GE, Yip C-M, Donovan BJ. Post-prostatectomy in-

continence: the influence of bladder dysfunction. J
Urol 1987; 138:574-578.

76. Mayo ME, Ansell JS. Urodynamic assessment of incon-

tinence after prostatectomy. J Urol 1979; 122:60-61.

77. Fitzpatrick JM, Gardiner RA, Worth PHL. The evalua-

tion of 68 patients with postprostatectomy inconti-
nence. Br J Urol 1979; 51:552-555.

78. Dorflinger T, Frimodt-MoUer PC, Bruskewitz RC,

512

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Jensen KM-E, Iversen P, Madsen PO. The significance
of uninhibited detrusor contractions in prostatism. J
Urol 1985; 133:819-821.

79. Abrams PH. Prostatism and prostatectomy: the value of

urine flow rate measurement in the preoperative as-
sessment for operation. J Urol 1977; 117:70-71.

80. Barkin M, Dolfin D, Herschorn S, Comisarow R, Fisher

R. Voiding dysfunction in institutionalized elderly
men: the influence of previous prostatectomy. J Urol
1983; 130:258-259.

81. Kimche D, Saar M, Lask D. Evoked response studies in

detrusor hyperreflexia due to infravesical obstruction
in neurological patients. J Urol 1985; 133:641-643.

82. Bradley WE. Neurologic disorders affecting the urinary

bladder. In: Krane RJ, Siroky ME, eds. Clinical neuro-
urology. Boston: Little Brown, 1979:245.

83. Khan Z, Hertanu J, Yang WC, Melman A, Leiter E. Pre-

dictive correlation of urodynamic dysfunction and
brain injury after cerebrovascular accident. J Urol
1981; 126:86-88.

84. Khan Z, Starer P, Yang WC, Bhola A. Analysis of void-

ing disorders in patients with cerebrovascular acci-
dents. Urology 1990; 35:265-270.

85. Borrie MJ, Campbell AJ, Caradoc-Davies TH, Spears

GFS. Urinary incontinence after stroke: a prospective
study. Age Ageing 1986; 15:177-181.

86. Tsuchida S, Noto H, Yamaguchi O, Itoh M. Urodynamic

studies on hemiplegic patients after cerebrovascular
accident. Urology 1983; 21:315-318.

87. Kitada S, Ikei Y, Hasui Y, Nishi S, Yamaguchi T, Osada

Y. Bladder function in elderly men with subclinical
brain magnetic resonance imaging lesions. J Urol
1992; 147:1507-1509.

88. Wein AJ, Raezer DM. Physiology of micturition. In:

Krane RJ, Siroky MB, eds. Clinical neuro-urology.
Boston: Little Brown, 1979:1-33.

89. Brocklehurst JC, Andrews K, Richards B, Laycock PJ.

Incidence and correlates of incontinence in stroke pa-
tients. J Am Geriatr See 1985; 33:540-542.

90. Aminoff MJ, Wilcox CS. Assessment of autonomic func-

tion in patients with a parkinsonian syndrome. Br
Med J 1971; 4:80-84.

91. Berger Y, Blaivas JG, DeLa Rocha ER, Salinas JM. Uro-

dynamic findings in Parkinson's disease. J Urol 1987;
138:836-838.

92. Andersen JT. Disturbances of bladder and urethral func-

tion in Parkinson's disease. Int Urol Nephrol 1985;
17:35-41.

93. Greenberg M, Gordon HL, McCutchen JJ. Neurogenic

bladder in Parkinson's disease. South Med J 1972; 65:
446-448.

94. Andersen JT, Bradley WE. Cystometric, sphincter and

electromyelographic abnormalities in Parkinson's dis-
ease. J Urol 1976; 116:75-78.

95. Fitzmaurice H, Fowler CJ, Rickards D, et al. Micturition

disturbance in Parkinson's disease. Br J Urol 1985;
57:652-656.

96. Galloway NTM. Urethral sphincter abnormalities in

parkinsonism. J Urol 1983; 55:691-693.

97. Pavlakis AJ, Siroky MB, Goldstein I, Krane RJ. Neu-

rourologic findings in Parkinson's disease. J Urol
1983; 129:80-83.

98. Sotolongo JR. Voiding dysfunction in Parkinson's dis-

ease. Semin Neurol 1988; 8(2):166-169.

99. Elliott JS. Urologic problems of the elderly patient. In:

Ebaugh FG, ed. Management of common problems in

geriatric medicine. Menlo Park, CA: Addison- Wesley,
1981:353.

100. Resnick NM, Yalla SV. Management of urinary inconti-

nence in the elderly. N Engl J Med 1985; 313:800-805.

101. Kaplan SA, Blaivas JG. Practical approach to the diag-

nosis of urinary incontinence. Semin Neurol 1988; 8:
131-136.

102. Schiff HI. The neurogenic bladder in diabetes. NY State

J Med 1982; 82:922-926.

103. Watkins PJ. The bladder in diabetes. Practitioner 1975;

214:56-59.

104. Frimodt-Moller C. Diabetic cystopathy. I. A clinical

study of the frequency of bladder dysfunction in dia-
betics. Dan Med Bull 1976; 23:267-278.

105. Leach GE, Yip CM. Urologic and urodynamic evaluation

of the elderly population. Clin Geriatr Med 1986; 2:
731-755.

106. Ellenberg M. Development of urinary bladder dysfunc-

tion in diabetes mellitus. Ann Intern Med 1980;
92(Part 2):321-323.

107. Motzkin D. The significance of deficient bladder sensa-

tion. J Urol 1968; 100:445-450.

108. Frimodt-Moller C, Mortensen S. Treatment of diabetic

cystopathy. Ann Intern Med 1980; 92(Part 2):327-328.

109. Andersen JT, Bradley WE. Abnormalities of bladder in-

nervation in diabetes mellitus. Urology 1976; 7:442-
448.

110. Buck AC, Reed PI, Siddiq YK, Chisholm GD, Eraser TR.

Bladder dysfunction and neuropathy in diabetes. Dia-
betologia 1976; 12:251-258.

111. Booth CM, Harrison MJG, Green JF. Evaluation of

screening tests for detecting sacral autonomic neurop-
athy in diabetes mellitus. Br J Urol 1984; 56:31-34.

112. Fagerberg SE, Kock NG, Petersen I, Stener I. Urinary

bladder disturbances in diabetics. I. A comparative
study of male diabetics and controls aged between 20
and 50 years. Scand J Urol Nephrol 1967; 1:19-27.

113. Frimodt-Moller C. Diabetic cystopathy. II. Relationship

to some late-diabetic manifestations. Dan Med Bull
1976; 23:279-287.

114. Starer P, Libow L. Cystometric evaluation of bladder

dysfunction in elderly diabetic patients. Arch Intern
Med 1990; 150:810-813.

115. Bradley WE. Diagnosis of urinary bladder dysfunction in

diabetes mellitus. Ann Intern Med 1980; 92(Part 2):
323-326.

116. Blaivas JG. Neurologic dysfunctions. In: Yalla SV,

McGuire EJ, Elbadawi A, Blaivas JG, eds. Neurourol-
ogy and urodynamics: principles and practice. New
York: Macmillan, 1988:349.

117. Rubinow DR, Nelson JC. Tricyclic exacerbation of undi-

agnosed diabetic uropathy. J Clin Psychiatry 1982; 43:
210-212.

118. Resnick NM, Yalla SV. Detrusor hyperactivity with im-

paired contractile function: an unrecognized but com-
mon cause of incontinence in elderly patients. JAMA
1987; 257:3076-3081.

119. Keegan GT, McNichols DW. The evaluation and treat-

ment of urinary incontinence in the elderly. Surg Clin
North Am 1982; 62:261-274.

120. Hilton P, Stanton SL. Algorithmic method for assessing

incontinence in elderly women. Br Med J 1981; 282:
940-942.

121. Resnick NM. Urinary incontinence in the elderly. Med

Grand Rounds 1984; 3:281-290.

122. Bruskewitz RC, Iversen P, Madsen PO. Value of postvoid

Vol. 60 No. 6

URINARY INCONTINENCE IN THE ELDERLY— ST ARER

513

residual urine determination in evaluation of prostat-
ism. Urology 1982; 20:602-604.

123. Turner-Warwick R. Some clinical aspects of detrusor

dysfunction. J Urol 1975; 113:539-544.

124. Diokno AC, Koff SA, Andersen W. Combined cystometry

and perineal electromyography in the diagnosis and
treatment of neurogenic urinary incontinence. J Urol
1976; 115:161-163.

125. Starer P, Libow LS. The measurement of residual urine

in the evaluation of incontinent nursing home resi-
dents. Arch Gerontol Geriatr 1988; 7:75-81.

126. Farrar DJ, Osborne JL, Stephenson TP, et al. A urody-

namic view of bladder outflow obstruction in the fe-
male: factors influencing the results of treatment. Br J
Urol 1976; 47:815-822.

127. Fossberg E, Sander S, Beisland HO. Urinary inconti-

nence in the elderly: a pilot study. Scand J Urol Neph-
rol 1981; 60(Suppl):51-53.

128. Sabanathan K, Duffin HM, Castleden CM. Urinary tract

infection after cystometry. Age Aging 1985; 14:291-
295.

129. Ouslander JG, Uman GC. Urinary incontinence: oppor-

tunities for research, education, and improvements in
medical care in the nursing home setting. In: Schnei-
der EL, Wendland CL, Zimmer AW, et al, eds. The
teaching nursing home. New York: Raven Press, 1985:
173-196.

130. Ouslander J, Leach G, Abelson S, Staskin D, Blaustein J,

Raz S. Simple versus multichannel cystometry in the
evaluation of bladder function in an incontinent geri-
atric population. J Urol 1988; 140:1482-1486.

131. Ouslander J, Staskin D, Raz S, Su H-L, Hepps K. Clin-

ical versus urodynamic diagnosis in an incontinent ge-
riatric female population. J Urol 1987; 137:68-71.

132. Sutherst JR, Brown MC. Comparison of single and mul-

tichannel cystometry in diagnosing bladder instabil-
ity. Br Med J 1984; 288:1720-1722.

133. Castleden CM, George CF, Renwick AG, Asher MJ.

Imipramine — a possible alternative to current therapy
for urinary incontinence in the elderly. J Urol 1981;
125:318-320.

134. Merrill DC, Bradley WE, Markland C. Air cystometry.

II. A clinical evaluation of normal adults. J Urol 1972;
108:85-88.

135. International Continence Society. Fourth report on the

standardisation of terminology of lower urinary tract
function. Br J Urol 1981; 53:333-335.

136. Drach GW, Layton TN, Binard WJ. Male peak urinary

flow rate: relationships to volume voided and age. J
Urol 1979; 122:210-214.

137. Drach GW, Ignatoff J, Layton T. Peak urinary flow rate:

observations in female subjects and comparison to
male subjects. J Urol 1979; 122:215-219.

138. Chancellor MB, Blaivas JG, Kaplan SA, Axelrod S. Blad-

der outlet obstruction versus impaired detrusor con-
tractility: the role of uroflow. J Urol 1991; 145:810-
812.

139. Silk E. Young and still growing — disposable diaper/adult

pad markets. Nonwovens Indust 1983; 14:32.

140. Prinsley DM, Cameron KP. Management of urinary in-

continence. Med J Aust 1979; 1:578-579.

141. Beber CR. Freedom for the incontinent. Am J Nurs 1980;

80:482^84.

142. Champlin L. Diaper market up. Med Prod Sales 1982;

13:69-77.

143. Castleden CM, DufTin HM. Guidelines for controlling

urinary incontinence without drugs or catheters. Age
Aging 1981; 10:186-190.

144. Ouslander JG, Kane RL. The costs of urinary inconti-

nence in nursing homes. Med Care 1984; 22:69-79.

145. Schnelle JF, Newman DR, Fogarty T. Management of

patient continence in long-term care nursing facilities.
Gerontologist 1990; 30:373-376.

146. National Institutes of Health Consensus Development

Conference: Urinary incontinence in adults. JAMA
1989; 261:2685-2690.

147. Stanton SL. Suprapubic approaches for stress inconti-

nence in women. J Am Geriatr Soc 1990; 38:348-351.

148. Ek A, Andersson K-E, Gullberg B, Ulmsten U. The ef-

fects of long-term treatment with norephedrine on
stress incontinence and urethral closure pressure pro-
file. Scand J Urol Nephrol 1978; 12:105-110.

149. Kinn A-C, Lindskog M. Estrogens and phenylpropanol-

amine in combination for stress urinary incontinence
in post menopausal women. Urology 1988; 32:273—
280.

150. Wells TJ. Pelvic (floor) muscle exercise. J Am Geriatr

Soc 1990; 38:333-337.

151. McDowell BJ, Burgio KL, Dombrowski M, Lochner JL,

Rodriguez E. An interdisciplinary approach to the as-
sessment and behavioral treatment of urinary incon-
tinence in geriatric outpatients. J Am Geriatr Soc
1992; 40:370-374.

152. Burgio KL, Engel BT. Biofeedback-assisted behavioral

training for elderly men and women. J Am Geriatr Soc
1990; 38:338-340.

153. Burgio KL, Whitehead WE, Engel BT. Urinary inconti-

nence in the elderly. Bladder-sphincter biofeedback
and toileting skills training. Ann Intern Med 1985;
104:507-515.

154. Hadley EC. Bladder training and related therapies for

urinary incontinence in older people. JAMA 1986; 256:
372-379.

155. Burton JR, Pearce KL, Burgio KL, Engel BT, Whitehead

WE. Behavioral training for urinary incontinence in
elderly ambulatory patients. J Am Geriatr Soc 1988;
36:693-698.

156. Fantl JA, Wyman JF, McClish DK, et al. Efficacy of

bladder training in older women with urinary incon-
tinence. JAMA 1991; 265:609-613.

157. Lamhut P, Jackson TW, Wall LL. The treatment of uri-

nary incontinence with electrical stimulation in nurs-
ing home patients: a pilot study. J Am Geriatr Soc
1992; 40:48-52.

158. Castleden CM, Duffin HM, Gulati RS. Double-blind

study of imipramine and placebo for incontinence due
to bladder instability. Age Aging 1986; 15:299-303.

159. Moisey CU, Stephenson TP, Brendler CB. The urody-

namic and subjective results of treatment of detrusor
instability with oxybutynin chloride. Br J Urol 1980;
52:472-475.

160. Ouslander JG, Blaustein J, Connor A, Orzeck S, Yong

CL. Pharmacokinetics and clinical effects of oxybuty-
nin in geriatric patients. J Urol 1988; 140:47-50.

161. Ouslander JG, Blaustein J, Connor A, Pitt A. Habit

training and oxybutynin for incontinence in nursing
home patients: a placebo-controlled trial. J Am Geri-
atr Soc 1988; 36:40-46.

162. Zorzitto ML, Holliday PJ, Jewett MAS, Herschorn S,

Fernie GR. Oxybutynin chloride for geriatric urinary
dysfunction. A double-blind placebo-controlled study.
Age Aging 1989; 18:195-200.

514

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

163. Diokno AC, Lapides J. Oxybutynin: a new drug with

analgesic and anticholinergic properties. J Urol 1972;
108:307-309.

164. Brendler CB, Radebaugh LC, Mohler JL. Topical oxybu-

tynin chloride for relaxation of dysfunctional bladders.
J Urol 1989; 14:1350-1352.

165. Castleden CM, Duflln HM, Asher MJ, Yeomanson CW.

Factors influencing outcome in elderly patients with
urinary incontinence and detrusor instability. Age Ag-
ing 1985; 1:303-307.

166. Hedlund H, Mattiasson A, Andersson K-E. Effects of

pinacidil on detrusor instability in men with bladder
outlet obstruction. J Urol 1991; 146:1345-1347.

167. Castleden CM, DufTin HM. Guidelines for controlling

urinary incontinence without drugs or catheters. Age
Aging 1981; 10:186-190.

168. Pannill FC, Williams TF, Davis R. Evaluation and treat-

ment of urinary incontinence in long-term care. J Am
Geriatr Soc 1988; 136:902-910.

169. Schnelle JF, Traughber B, Morgan DB, Embry JE, Bin-

ion AF, Coleman A. Management of geriatric inconti-

nence in nursing homes. J Appl Behav Anal 1983; 16:
235-241.

170. Schnelle JF, Traughber B, Sowell VA, Newman DR, Pet-

rilli CO, Ory M. Prompted voiding treatment of uri-
nary incontinence in nursing home patients. A behav-
ior management approach for nursing home staff. J
Am Geriatr Soc 1989; 37:1051-1057.

171. Hu T-W, Igou JF, Kaltreider DL, et al. A clinical trial of

a behavioral therapy to reduce urinary incontinence in
nursing homes. Outcome and implications. JAMA
1989; 261:2656-2662.

172. Engel BT, Burgio LD, McCormick KA, Hawkins AM,

Scheve AAS, Leahy E. Behavioral treatment of incon-
tinence in the long-term care setting. J Am Geriatr
Soc 1990; 38:361-363.

173. Schnelle JF, Sowell VA, Hu T-W, Traughber B. Reduc-

tion of urinary incontinence in nursing homes: does it
reduce or increase costs? J Am Geriatr Soc 1988; 36:
34-39.

174. O'Donnell PD, Finkbeiner AE, Beck C. Urinary inconti-

nence volume measurement in elderly male inpa-
tients. Urology 1990; 35:499-503.

Falls in Older Persons

Rein Tideiksaar, Ph.D.

Falls are a common problem for older persons.
Studies indicate that up to one-third of commu-
nity-dwelling persons over age 65 fall annually
(1, 2), and approximately one-half of these have
more than one fall (1). The prevalance of falls is
equally problematic among older persons in insti-
tutional settings. Falls in the acute-care hospital
are a leading cause of adverse effects, and older
patients experience the overwhelming majority of
falls (3), as many as 10% of these older hospital
patients falling repeatedly (4). About 50% of
nursing home residents fall every year (5), and up
to 40% have recurrent falls (5). The risk of falling
increases with age. The fall rate (number of falls
per 100 person-years) increases from 47 for those
aged 70-74 to 121 for those aged 80 and older (2).
The fall rate is equal for men and women (2).

Within the next few decades the older popu-
lation— especially those persons aged 75 and
older, who are at greatest risk of falling — will in-
crease, and physicians in all specialties will be
faced with and challenged by the problem of falls
in older people. The purpose of this article is to
review the consequences of falls, the conditions
under which falls occur, the factors associated
with risk of falling, and intervention strategies
designed to reduce that risk.

Consequences of Falls

Falls are associated with significant morbid-
ity and mortality. Falls are a leading cause of
mortality due to "accidents" or unintentional in-

Adapted from Preventing Frailty: Osteoporosis and Falls, a
symposium presented at The Mount Sinai Medical Center,
New York, NY, on Oct. 30, 1992.

From the Department of Geriatrics and Adult Develop-
ment, The Mount Sinai Medical Center, and The Jewish Home
and Hospital for Aged, New York, New York. Address reprint
requests to Rein Tideiksaar, Ph.D., Department of Geriatrics
and Adult Development, Box 1070, The Mount Sinai Medical
Center, One Gustave Levy Place, New York, NY 10029.

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

jury, with up to 9,500 deaths per year attributed
to falling (6). For those surviving a fall, signifi-
cant morbidity is likely to follow. About 5% of
falls of all older persons, whether living in their
community or institutionalized, result in a frac-
ture (1). Hip fractures are the most serious. After
repair of a hip fracture, many persons never regain
their previous level of ambulation. Approxi-
mately 60% of persons experience decreased mo-
bility, and another 25% become functionally
dependent in ambulation and require either me-
chanical (cane, walker) or human assistance in
walking (7). Women have a higher rate of Hip
fracture than men, partly owing to low bone mass
(8). Osteoporosis can either result in spontaneous
hip fractures or lead to fractures that occur from
the impact of a fall onto hard, nonabsorbing
ground surfaces. An additional 10% of falls result
in nonfracture injury (1), including hematomas,
joint dislocations, muscle sprains, and head inju-
ries.

In the absence of physical injury, falls are by
no means innocuous. Up to 50% of those who have
had a fall admit to avoiding activities, typically
those that led to the fall, because of a fear of fur-
ther falls or injury (9). The risk of developing a
fear of falling is increased in those persons who
have underlying gait and balance disorders and
who have recurrent falls over a short period; the
risk is greatest in those who suffer injury, func-
tional decline, or prolonged postfall lie times —
inability to get up from the floor by oneself Any
immobility that occurs as a result of injury or fear
of falling places persons at risk for a host of com-
plications, including deep venous thrombosis and
pulmonary embolism; sepsis arising from decu-
bitus ulcers, urinary tract infections, and hy-
postatic pneumonias; muscle atrophy; and joint
contractures. The likelihood of incurring a com-
plication is in direct proportion to the duration of
immobility.

Falls experienced by older persons are also

515

516

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

distressing for family members. At issue is
whether the person can live at home alone and
remain safe or requires protective nursing home
placement. Falls and instability are a contribut-
ing factor in up to 409c of nursing home admis-
sions (6).

Causes of Falling

Falls are likely to occur during any activity
that results in balance displacement (walking,
climbing and descending stairs, transferring onto
or off chair or bed) and in which the body mech-
anisms responsible for controlling stability fail.
In older persons, the catalysts for balance loss and
subsequent falls are intrinsic factors, such as age-
related physiologic changes, pathologic condi-
tions, and medication effects, and extrinsic fac-
tors or environmental hazards.

Among the elderly, the causative factors for
falls vary widely. For fit and active persons, a fall
may be "accidental," the consequence of adjusting
to age-related changes (such as a decline in visual
function) in interaction with unfavorable envi-
ronmental conditions (such as the presence of an
uneven ground surface). Failure to recognize a
curb edge or door sill in one's path can lead to
tripping and the risk of falling. However, the on-
set of falling is more serious for frail persons in
poor health. For these individuals, falls are typi-
cally a sign or symptom of underlying pathologic
conditions, adverse drug effects, and unsafe envi-
ronmental conditions, factors that may be present
either separately or in combination to precipitate
a fall.

Age-Related Physiologic Factors

A number of physiologic changes occur with
age and contribute to postural stability and fall
risk. The most important changes occur in the
visual and neuromuscular systems, affecting gait
and balance and altering a person's ability to ma-
neuver about the environment safely.

Vision, Visual function, the ability of the
eyes to adjust to various environmental stimuli,
diminishes with age. Because the response to
varying levels of light and darkness is reduced
(10), persons require more time to adapt to
changes in lighting, for example in moving from
areas of bright illumination to areas of darkness.
Dark adaption, the capability of the eye to adjust
to low levels of illumination, is particularly af-
fected by age and is associated with the risk of
falling (11). This change compromises the ability
to recognize potential environmental hazards un-
der conditions of low illumination. Bright light-

ing produces similar effects. With age, pupillary
response and accommodation declines (10), and
consequently, excessive illumination may lead to
momentary blindness until the eyes adjust to it.

A greater sensitivity of the aging eye to
glare — a dazzling effect associated with a source
of intense illumination — can interfere with vision
and increase fall risk. Common sources of glare
include sunlight and unshielded or exposed light
bulbs. These sources can also result in reflected
glare off polished floor surfaces, thus hiding from
view potential ground hazards that can lead to
slips and trips. Glare on floor surfaces is per-
ceived by older persons as slipperiness. To com-
pensate, they may alter their gait, assuming a
pattern that is slower and flat-footed, with a wide
base of support. However, such adaptions may not
be possible for persons with neuromuscular ab-
normalities and, when attempted, may lead to in-
stability and falls.

Depth perception also declines with age (10).
As a result, environmental objects of low visual
contrast and thus indistinguishable from their
background, such as carpet and step edges and
door sills, may not be easily visualized. Pat-
terned, checkered, or floral floor tiles and carpet
designs may appear to the aging eye as having
either ground elevations or depressions, possibly
leading to hazardous gaits.

Balance and Gait. Maintaining balance is
dependent on correct functioning of the sensory
and motor or musculoskeletal systems. Within
the sensory system, proprioception, vestibular in-
put, and vision are the most critical. Operating in
unison, these components culminate in postural
sway, an anteroposterior and lateral motion of
the body that counters the effects of gravity and
controls stability. Proprioceptive feedback, aris-
ing from muscle, tendon, and joint receptors in
the limbs and neck, provides the body with kin-
esthetic information and orientation with respect
to environmental conditions. The vestibular sys-
tem detects linear and angular displacements of
the head (via receptors located in the semicircular
canals and otolith organs) and, in response to dis-
placements of balance, initiates a righting reflex
that consists of a series of compensatory limb,
trunk, and head movements that control stabil-
ity. The visual system provides the majority of
the sensory input needed for gait and balance.
Visual input helps one to perceive cues within the
surrounding environment — to detect altered lin-
ear and angular motions of the visual field — and
supplies the body with information on its orien-
tation in space. In addition, vision facilitates
proprioceptive and vestibular feedback, and, in

Vol. 60 No. 6

FALLS IN OLDER PERSONS-TIDEIKSAAR

517

the event of any deficiencies, vision serves as a
substitute to preserve balance.

The musculoskeletal system provides pos-
tural support (proper body alignment), muscular
strength, and joint flexibility to maintain balance
and correct body displacements. When standing,
a person maintains balance by constantly posi-
tioning the body's center of gravity (COG) over a
base of support (BOS), the area surrounding the
feet, by utilizing an ankle strategy that consists
of rotating the body about the ankle joints. Dur-
ing ambulation, the COG extends beyond the
BOS and stretches the limits of stability. This dis-
placement is detected by the sensory systems,
which, in turn, send a group of messages to
stretch receptors in the joints and muscles of the
lower extremities. To regain stability through re-
alignment of the COG over the BOS, a series of
postural maneuvers are initiated. Depending on
the size of the displacement, either a hip strategy
consisting of flexing or extending the hips or a
stepping or stumbling strategy consisting of a
rapid forward or backward shifting movement of
the feet is employed, in conjunction with a right-
ing reflex or response.

Also, gait — the manner of walking — is pre-
dominantly a function of the musculoskeletal sys-
tem. The gait cycle consists of a stance and a
swing phase. Ambulation is brought about by a
series of alternating leg movements: pushing off
on the leg in stance while, at the same time,
swinging the other leg forward. To allow for ade-
quate ground clearance during the swing phase,
the knee becomes flexed and the ankle dorsi-
flexed. When the heel of the swing leg strikes the
ground, a return to stance, the knee becomes ex-
tended and the foot plantar-flexed to provide body
support.

Any disruption in the effective coordination
of sensory and musculoskeletal movements jeop-
ardizes postural stability and gait and increases
the likelihood of falls. With age, proprioceptive
feedback declines, and its malfunction is associ-
ated with increased postural sway to a level of
unsteadiness (12). Vision can compensate for
proprioceptive loss and support balance, as dem-
onstrated by older persons with proprioceptive
loss who ambulate by visually observing the
ground to ensure correct foot placement. How-
ever, when visual input is decreased, maintaining
balance becomes difficult, as evidenced when a
person is asked to stand with eyes closed or to
walk in a dark room; sway increases and balance
becomes unsteady. Also, a decline in visual per-
ception may alter the processing of visual infor-
mation and predispose to instability. Older per-

sons who fall have more difficulty than nonfallers
in estimating true horizontal and vertical planes
in their visual fields (13). In addition, dependence
on visual input for balance is often inadequate
because visual feedback to control stability is of-
ten slow, and any correction for change is delayed
in older persons (14).

The vestibular righting response also dimin-
ishes with age (15). As a result, if a person trips or
slips — experiences a displacement of balance —
the chances of recovering stability decline.
Within the musculoskeletal system, posture be-
comes more kyphotic, related, in part, to osteopo-
rosis and muscular weakness (16), possibly alter-
ing an individual's balance threshold because the
COG is shifted forward, extending beyond the
BOS. In addition, muscular strength in the legs
and ankles declines, and joint flexibility becomes
impaired with age (17), impairing the effective
application of ankle, hip, and stepping strategies
to correct balance.

Changes in musculoskeletal function contrib-
ute to a number of gait alterations. Compared to
younger persons, the gait of older persons is de-
creased in speed, stride length (the distance the
foot travels during the swing phase), heel lift (the
level of ground clearance by the foot during the
swing phase), and the ability of the foot to fully
plantarflex and dorsiflex (18). As a consequence,
gait becomes less steady, and the execution of mo-
tor strategies to adjust balance displacements is
impaired.

Although older persons normally are slow in
detecting and responding to postural distur-
bances, there is a good deal of reserve and redun-
dancy in the sensory information and motor func-
tion needed to maintain gait and balance. Failure
of one source of input can be compensated for by
feedback from another system; for example, vi-
sual input can compensate for a decline in pro-
prioceptive feedback. Similarly, when perturbed,
older persons with impairments are able to resort
to alternative motor strategies. If ankle strate-
gies cannot be employed because of lower-extrem-
ity weakness, older persons are able to compen-
sate by using a hip strategy — a proximal-to-distal
sequence of thigh and trunk muscle contrac-
tions— to preserve balance. However, dysfunction
in more than one system is likely to result in a
lowered balance threshold and increased fall risk.

Pathologic Conditions

Acute and chronic medical problems that af-
fect gait and balance are more decisive in causing
falls than age-related physiologic changes not ac-

518

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

companied by medical problems. It has repeatedly
been demonstrated that fallers tend to have more
medical diagnoses than nonfallers (1). The re-
sponsible acute diseases include any condition
that interferes with postural stability, such as
syncope, hypovolemia, cardiac arrhythmias, elec-
trolyte disturbances, stroke, and sepsis (urinary
tract infections, gallbladder disease, pneumo-
nias).

Chronic diseases that may exacerbate pos-
tural instability and lead to gait impairments
originate in the visual and neuromuscular sys-
tems. For example, diseases of the eye, such as
cataracts, macular degeneration, and glaucoma,
significantly interfere with visual function and
are strongly associated with both the risk of falls
and hip fractures (9, 19). Impaired vision becomes
an even greater risk factor when combined with
poor environmental lighting conditions.

Parkinson's disease is associated with a loss
of the postural reflexes of propulsion (uncon-
trolled forward motion of the body) and retropul-
sion (producing loss of balance backward). Per-
sons with Parkinson's disease also exhibit a
number of gait changes: walking becomes short-
stepped and shuffling, barely clearing the floor;
persons typically display poor initiation and hes-
itancy or freezing of gait. Lower extremity hemi-
plegia or paresis results in a diminished BOS and
decreased ankle dorsiflexion on foot clearance
when walking.

Arthritis of the hips and knees can lead to
decreased single limb support (the result of pain
when bearing weight), altered gait (reduced
stride length, ground clearance), and instability
(unstable and inflexible joints resulting in poor
corrective responses). Neuropathy, osteomalacia,
thyroid disease, and deconditioning can lead to
lower extremity muscle weakness, contributing
to altered gait and balance; for example, gluteal
muscle weakness results in exaggerated trunk
movements and waddling gait. Foot abnormali-
ties such as bunions, hammertoes, or heel spurs
can lead to mechanical gait impairments.

Alzheimer's disease is associated with poor
gait (reduced proprioception and steppage
height), altered visual input (visual spatial and
field defects), and impaired judgment (inability to
distinguish between safe and unsafe activities
and environmental conditions).

Medications

Certain medications are known to interfere
with postural stability and gait and to increase
fall liability. The most common include sedatives.

antipsychotics, antihypertensives, and antide-
pressants (6, 9). These can lead to increased seda-
tion, delayed reaction times, orthostasis, move-
ment disorders, and cognitive impairment such as
diminished awareness of hazards. The risk of
drug-related falls increases with long-acting med-
ications and polypharmacy (taking more than
four drugs simultaneously) (1).

Extrinsic Factors

The overwhelming majority of falls, whether
in the community or in institutional settings,
take place in the bedroom and bathroom (20).
This may reflect the increased time older persons
spend in these locations or, more important, may
represent difficulties in the performance of mobil-
ity tasks. Several environmental obstacles and
design features are associated with falling: too
low or too high beds and elevated side rails; low
chair seats and poor armrest support; low-seated
toilets and poor grab-bar support; poorly illumi-
nated areas; elevated ground conditions (door
thresholds, thick pile carpeting); and slippery
ground surfaces (wet or polished floors, non-slip-
resistant bathtub surfaces, sliding rugs) (20). The
type of footwear, including fit, thickness and slip
resistance of soles, heel height, and the condition
and proper utilization of assistive devices (canes,
walkers) also influence gait and balance. The
likelihood of environmental conditions contribut-
ing to falls is greatest in those persons with al-
tered mobility. Often, the presence of environ-
mental obstacles exceeds the person's functional
ability to walk and transfer safely and interferes
with their ability to compensate.

Interventions

The goals of fall prevention are to maximize
mobility, reduce the threat of injury, and main-
tain autonomy. Interventions designed to accom-
plish these goals encompass medical, rehabilita-
tive, and environmental approaches.

Medical Interventions. The medical ap-
proach is twofold: first, to identify persons at risk
of falls and, for those with a history of recent falls,
to uncover the causative factors; second, to at-
tempt to modify any discovered risk factors.

The identification of fall risk is determined
by reviewing the person's medical history for con-
tributory conditions and medications, including a
history of recent falls, lower extremity dysfunc-
tion, arthritis of the hips and knees, gait and bal-
ance impairment, altered cognition, visual disor-
ders, postural hypotension, bladder dysfunction
(nocturia, incontinence), and use of certain med-

Vol. 60 No. 6

FALLS IN OLDER PERSONS— TIDEIKSAAR

519

ications (psychotropics, sedatives, hypnotics, an-
tihypertensives) (5, 6).

In the absence of historical risk factors, al-
tered mobility is a strong indicator of fall risk
(21). Mobility dysfunction is ascertained by ask-
ing the person to perform a number of gait and
balance maneuvers. The person should be ob-
served sitting and rising from a chair, standing
with eyes closed (the Romberg test to assess prop-
rioceptive function), maintaining postural stabil-
ity when perturbed (this is a sternal push test
accomplished by lightly pushing on the sternum
and initiating displacement), and walking in a
straight line and turning around. An inability to
perform one or more of these maneuvers indicates
a risk of falling. Moreover, an abnormal finding
signifies the presence of an underlying intrinsic
or extrinsic problem (Table 1). The presence of
historical risk factors or mobility dysfunction
should trigger further investigations aimed at
modifying the responsible factors.

In those persons who have had falls, the eval-
uation begins with a history of the surrounding
circumstances of the event(s). Inquiries need to be
made about symptoms experienced, location of
the fall, activity performed at the time, and time
or hour of the episode(s). This information pro-
vides important clues to causative factors. For
example, an individual who has had one or more
falls who complains of dizziness and falls during
transfers in or out of bed or rising from a chair
may have orthostatic hypotension.

Once the history of falls is obtained, past
medical problems, current complaints, and medi-
cations should be reviewed. A physical examina-
tion and diagnostic studies should follow, the ex-
tent of both determined by the historical
information gathered and clinical suspicions. The
goal of the medical evaluation is to rule out con-
tributing acute and chronic disease processes and
medications and to treat each accordingly.

Rehabilitative Interventions. Those persons
who fail to improve with medical treatment and
continue to remain at risk of falling may respond
to a number of rehabilitative strategies. For those
with chronic neuromuscular disorders, a focused
program of exercises aimed at correcting gait and
balance impairments may be beneficial in reduc-
ing risk (22). For example, in persons with degen-
erative arthritis, stretching exercises directed
toward enhancing joint flexibility and strength-
ening the hip and knee extensor and flexor mus-
cles can improve mobility. For persons with Par-
kinson's disease, ambulation training, proximal
and ankle muscle strengthening, and postural ex-
ercises can enhance both gait and balance. For

TABLE 1

Differential Diagnoses of Mobility Dysfunction

Impaired
maneuver

Causes

Intrinsic

Extrinsic

Chair transfer

Standing
balance

Romberg test

Sternal nudge
test

Walking,
turning

Parkinsonism
Arthritis
Deconditioning
Postural

hypotension
Vestibular

dysfunction
Adverse drug

effects
Proprioceptive

dysfunction

Adverse drug

effects
Parkinsonism

Normal -pressure
hydrocephalus
Adverse drug

effects
Gait disorders

(parkinsonism,

hemiparesis,

foot problem)
Sensory

dysfunction

Adverse drug
effects

Poor chair design

Poor illumination
Overly absorptive

footwear or

carpeting

Improper footwear
Improper size or
use of ambula-
tion devices
Hazardous (slippery,
uneven) ground
surfaces

those with dementia, a daily program of walking
can offset altered gait and balance that commonly
results from immobility, and participation in a
regular program of exercise has been shown to
improve cognitive functioning and may reduce
falls that result from poor judgment, such as tak-
ing part in hazardous activities (22). In older
women with osteoporosis, weight-bearing exer-
cises can minimize further bone loss, thereby re-
ducing the risk of fracture. For those whose bone
density is already well below the fracture thresh-
old, exercise may be useful for its effect on prox-
imal muscle strength and coordination of gait and
balance, ultimately reducing fall risk.

Attention to footwear can support safe pat-
terns of gait. Shoes with rubber or crepe soles
provide adequate slip resistance on most ground
surfaces. Socks with nonskid treads are helpful
for persons who prefer to walk about barefooted.
High-heeled footwear should be avoided because
it decreases the standing and walking base of sup-
port and promotes loss of balance; shoe heels no
more than 3.5 cm in height and at least 5.5 cm in
surface width provide maximum balance support.

An ambulation device (cane, walker) for per-

520

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

TABLE 2

Environmental Hazards for the Elderly and Corrective Modifications

Hazard

Modification

Hazard

Modification

Illumination

Low lighting

Glare

Ground Surfaces
Slippery area rugs

Upended carpet edges

Slippery tiled or
linoleum kitchen
and bathroom floor
surfaces

Raised door sills

Furnishings
Furniture obstructs
pathways

Low seats on chairs

Too low or too high
bed height

Install good lighting in all
living areas, with switches at
room entrances; night lights
along route from bedroom to
bathroom; lamp next to bed;
stairways and steps
especially well lighted

Control window glare from
sunlight with Mylar or
translucent shades; they
eliminate glare but do not
reduce existing illumination

Eliminate glare from exposed
light bulbs by using
translucent light shields

Eliminate floor glare from
lights shining directly on
polished floor surfaces by
using carpeting or redirecting
light sources

Place slip-resistant matting

underneath rugs to firmly

anchor them
Tack or tape down (with

double-sided adhesive tape)

carpet edges
Apply nonskid adhesive floor

strips near areas that are

likely to become wet (sink,

toilet)

Place carpeting over raised sills
to create a smooth transition
surface between rooms

Rearrange furniture not to
protrude into natural
walkways

Supply chairs with armrests to
provide leverage and
compensate for low seat
height; add a cushion to raise
seat height

Replace existing mattress with
one either thinner or thicker
to lower or raise bed height

Shelves

Kitchen and closet
shelves too low or
too high for safe
reach

Stairways

Stairs lack handrails

Inadequate stairway
lighting

Stairway steps not

contrasting
Slippery stairway

steps

Bathroom

Slippery bathtub floor

Slippery bathtub rim

Slippery sink tops and
towel bars

Low toilet seats

Use either sturdy stools with
handrails or hand-held
grabber device, or place items
on reachable shelves
(between person's eye and hip
level)

Install well-anchored
cylindrical handrails (for
hand grasp)

Provide accessible day and
night lighting with switches
at top and bottom of stair

Apply color-contrasted adhesive
tape to step edges

Apply nonskid treads to all
steps

Apply nonskid strips or install
rubber mat; provide bath seat
and extended shower hose

Install portable grab bar to side
of tub

Apply nonskid adhesive tape to
sink tops (for hand grasp)
and replace towel bars with
grab bars

Use elevated toilet seats;
install grab bars on toilet
seat or wall

sons with underlying gait or balance disorders
improves mobility by increasing the person's
standing and walking base of support, providing
additional points of ground contact, and improves
stability by supplying proprioceptive feedback
through the handle. Such a device also provides a
sense of confidence and helps reduce the fear of
instability and falls. To ensure optimum function,
ambulation devices need to be "prescribed" — de-
signed for specific gait and balance problems and
tailored to fit the person and his or her environ-
mental setting.

Environmental Interventions. Environmen-
tal interventions aimed at reduction of falls con-
sist of identifying and eliminating hazardous con-
ditions that interfere with mobility. The best
method of determining whether a particular en-
vironmental area or furnishing is safe or hazard-
ous is to observe the person's mobility. Ask the
person to walk through every room and over floor
surfaces, including rugs, carpets, and tiled sur-
faces, to transfer on and off beds, chairs, and toi-
lets and in and out of bathtubs, to climb and de-
scend stairs, and to reach up and bend down to

Vol. 60 No. 6

FALLS IN OLDER PERSONS-TIDEIKSAAR

521

obtain objects from kitchen and closet shelves.
Those environmental features that interfere with
mobility can be altered (Table 2).

Summary

Falls are a common problem for older persons
and result in significant mortality and morbidity.
Most are caused by multiple intrinsic and extrin-
sic factors that have cumulative effects on mobil-
ity. Preventive interventions are aimed, first, at
identifying and assessing fall risk and, second, at
attempting to reduce those risks.

References

1. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls

among elderly persons living in the community. N Engl
J Med 1988; 319:1701-1707.

2. Campbell AJ, Borrie MJ, Spears GF, et al. Circumstances

and consequences of falls experienced by a community
population 70 years and over during a prospective
study. Age Ageing 1990; 19:136-141.

3. Jones WJ, Smith A. Preventing hospital incidents: what

we can do. Nurs Management 1989; 20(9):58-60.

4. Morgan VR, Mathison JH, Rice JC, et al. Hospital falls: a

persistent problem. Am J Public Health 1985; 75:775-
777.

5. Rubenstein LZ, Robbins AS, Schulmen BL, et al. Falls and

instability in the elderly. J Am Geriatr Soc 1988; 36:
266-278.

6. Tinetti ME, Speechley M. Prevention of falls among the

elderly. N Engl J Med 1989; 320:1055-1059.

7. Evans JG, Prudham D, Wandless I. A prospective study of

fractured proximal femur: incidence and outcome. Pub-
lic Health 1979; 93:235-241.

8. Cummings SR. Are patients with hip fractures more os-

teoporotic? A review of the evidence. Am J Med 1985;
78:487^93.

9. Nevitt MC, Cummings SR, Kidd D, et al. Factors for re-

current nonsyncopal falls: a prospective study. JAMA
1989; 261:2663-2668.

10. Kolanowski AM. The clinical importance of environmen-

tal lighting to the elderly. J Gerontol Nurs 1992; 18:
10-14.

11. McMurdo MET, Gaskell A. Dark adaption and falls in the

elderly. Gerontology 1991; 37(4):221-224.

12. Skinner HB, Barrack RL, Cook SD. Age-related declines

in proprioception. Clin Orthop 1984; 184:208-211.

13. Tobis JS, Reinsch S, Swanson JKM, et al. Visual domi-

nance of fallers among community-dwelling adults. J
Am Geriatr Soc 1985; 33:330-331.

14. Pykko I, Jantti P, Aalto H. Postural control in elderly

subjects. Age Ageing 1990; 19:215-221.

15. Norre ME, Forrez G, Beckers A. Post urography measur-

ing instability in vestibular dysfunctioning in the el-
derly. Age Ageing 1987; 16:89-93.

16. WoodhuU-Mcneal AP. Changes in posture and balance

with age. Clin Exp Res 1992; 4:219-225.

17. Davies CTM, Thomas DO, White MJ. Mechanical proper-

ties of young and elderly human muscle. Acta Med Scan
1986; 811(Suppl):219-226.

18. Sudarsky L. Gait disorders in the elderly. N Engl J Med

1990; 322:1441-1446.

19. Grisso J A, Kelsey J A, Strom BL, et al. Risk factors for

falls as a cause of hip fracture in women. N Engl J Med
1991; 324:1326-1331.

20. Tideiksaar R. Falling in old age: its prevention and treat-

ment. New York: Springer, 1989.

21. Tinetti ME, Ginter SF. Identifying mobility dysfunctions

in elderly patients. JAMA 1988; 259:1190-1193.

22. Elward K, Larson EB. Benefits of exercise for older adults.

Clin Geriatr Med 1992; 8(l):35-50.

Falls Prevention

The Efficacy of a Bed Alarm System in an Acute-Care Setting

Rein Tideiksaar, Ph.D., Clifford F. Feiner, D.O., and Jan Maby, D.O.

Abstract

The present study examined the clinical efficacy of a bed alarm system in reducing falls
from bed on a geriatric evaluation and treatment unit. A nine-month case-controlled study
was designed, in which 70 patients (60 women, 10 men; mean age 84 years, range 67-97
years) at increased risk for bed falls were randomly assigned to either an experimental or
a control group. Subjects in the experimental group (n = 35) received a bed alarm system
and those in the control group in = 35) did not. Outcome measures included bed falls,
performance of the bed alarm system, and staff attitudes toward the use of the system.
Although results failed to demonstrate a statistical difference in bed falls between the
experimental (n = 1) and control (n = 4) groups ip = 1.00), there was a clinical trend
toward reduced falls in the experimental group. The system functioned properly, activat-
ing an alarm in all instances when patients were transferring from bed, and with the
exception of one case, nurses could respond in a timely fashion to assist patients and
prevent bed falls. The system did not produce any adverse effects in patients, nor did the
device interfere with the rendering of medical care. The system was well accepted by
patients, families, and nurses. These data suggest that bed alarm systems are beneficial
in guarding against bed falls and are an acceptable method of preventing falls.

Patient falls are a leading cause of adverse
events in acute-care hospitals, accounting for up
to 40% of all incidence reports (1). The majority of
falls occur in patients 65 years of age and older
(2). Up to 75% take place during transfers out of
bed (3) and occur most frequently in patients with
diminished mobility and cognitive skills. Bed
falls in older patients are associated with a num-
ber of morbid complications, including soft tissue
injury, fractures, and immobility due to a fear of
further falls. Mechanical restraints are often em-
ployed to prevent falls; the prevalence of mechan-
ical restraints on general medical floors ranges
from 7% to 17% (4). Despite the use of such re-

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center (RT, CFF, JM), The Jewish
Home and Hospital for Aged (RT), and The Bronx Veterans
Administration Medical Center (JM). Address reprint re-
quests to Rein Tideiksaar, Ph.D., Department of Geriatrics
and Adult Development, Box 1070, The Mount Sinai Medical
Center, One Gustave Levy Place, New York, NY 10029.

straints, patients continue to experience bed falls
(5). As a result, bed alarm systems (BAS) have
been enlisted as alternative fall prevention mea-
sures.

Such devices warn nursing staff that patients
who should not be leaving their beds unassisted
are doing so. The bed alarm systems function by
allowing patients to maintain a free-movement
zone or area for normal activity in bed, including
turning around or rolling over. When patients
leave their beds and thus exceed the free-move-
ment zone, an alarm sounds, indicating that the
patient is about to transfer unsafely from bed.
While the use of these alarm systems to guard
against bed falls appears attractive, little re-
search on their performance and acceptance has
been published, aside from scant anecdotal re-
ports attesting to their benefits (6, 7). Therefore, a
case-control prospective study was undertaken to
determine the efficacy of one such system to re-
duce the number of bed falls in an acute-care hos-
pital setting.

522

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

PREVENTION OF FALLS-TIDEIKSAAR ET AL.

523

Methods

The study was conducted in the Geriatric
Evaluation and Treatment Unit of The Mount
Sinai Medical Center, New York City. The unit is
a 16-bed acute-care facility designed specifically
to provide multidisciplinary medical care to older
patients. All patients admitted to the unit be-
tween January 1 and September 30, 1992, were
evaluated by performance-oriented environmen-
tal mobility screen, an observational tool used to
assess bed mobility and the risk of falling (Fig.
1). Patients are observed performing a number of
transfer maneuvers and are scored as either nor-
mal or abnormal on each. One or more abnormal
scores signifies poor bed mobility and increased
fall risk.

Patients demonstrating poor bed mobility
were randomly assigned to either the experimen-
tal group that received an alarm system or the
control group that did not receive one. Informed
consent was obtained from all patients in the ex-
perimental group and their family members prior
to their being given an alarm. The unit staff were
not blinded to the intervention. Both groups were
given equal nursing attention to prevent falls:
hourly visual checks, mechanical restraints as
clinically indicated. The number of falls, the sur-
rounding circumstances of each, and any resul-
tant physical injury were recorded for both groups
during the course of their hospital stay. "Falls"
were defined as all events for which staff filed an
incident report indicating a fall. The difference in
number of falls from bed occurring in the experi-
mental and the control group served as an out-
come measure to evaluate the efficacy of the sys-
tem. Fisher's exact test was used to determine the
statistical significance of the difference.

The system used in this study was the RN +
OnCall bed monitoring system (Fig. 2), consist-
ing of a pressure-sensitive pad placed on top of the
patient's mattress and underneath the bed sheet.
In bed, the patient rests on the pressure pad. If
the patient sits up, the pressure on the sensor pad
is relieved, activating both an audio and a visual
alarm to a separate console located at the nursing
station.

Additional data collection for the experimen-
tal and control groups included diagnoses and
medications taken on admission and the use of
mechanical (physical) restraints and bed rails
during the patient's hospital stay. Data on system
performance included number of patient-use
hours; number of alarm activations; reasons for
alarm activation (why the patient was transfer-
ring out of bed); number of true alarms (patient

OBSERVA-nON

NORMAL

■ Bed transfer is
smooth, corTtroded
movement (sits on
and rises from bed in
one attempt)

■ Sitting balance
is stable

a Does not use arm
support to maintain
sitting balance

■ When seated both
feet rest flat on
grourxj

■ Able to lie down
(in supine position)
and rise In one snxx^th
controded movement

■ When seated 'eet
do not slide away
on ground

■ Able to operate
nurse call system

■ When seated bed does
not slide away

ABNORMAL

■ Bed transfer is not
smooth (requires several
attempts to sft or rise;
falls o" to mattress; uses
mattress edge to guide
transfers)

■ Sitting b>alance is
unstable

■ Uses arm support to
maintain sitting
t>alance

■ Feet do not rest
flat on ground

■ Unable to lie down

(in supine position) and
rise In one srrxx)th,
controlled movement

■ Feet slide away

■ Unable to operate
nurse call system

■ Bed slides away

■ Unable to perform
unassisted bed transfers

Fig. 1. Performance-oriented environmental mobility
screen.

was actually exiting the bed) and patient's loca-
tion (for example, sitting in bed, climbing over
side rails, or out of bed and walking about); num-
ber of false alarms (for instance, patient lying su-
pine in bed awake or asleep, but not getting out of
bed); average nurse response time per alarm ac-
tivation; presence of adverse affects (system in-
terfered with treatment or caused the patient
harm); and attitudes of patients, family members,
and nursing staff toward the system. This infor-
mation was collected and recorded by the nursing
staff on each shift.

Results

The experimental and the control group each
consisted of 35 patients (30 women; 5 men), mean
age 84 years and range between 67 and 97 years.
Patients had one or more acute medical problems
or underlying chronic neuromuscular conditions
(altered cognition, gait and balance disorders)

524

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Fig. 2. The RN + OnCall bed monitoring system.

and were taking medications (sedatives, antide-
pressants, antipsychotics) that contributed to
poor bed mobility either in isolation or in combi-
nation with inadequate bed design, including el-
evated bed height, unsupportive mattress, unsta-
ble bed wheels, slippery floor surfaces.

During the nine-month evaluation period, a
total of 24 falls occurred on the unit. The study
population had a total of 17 falls (Table 1). The
experimental group experienced a total of 5 falls:
1 fall from bed and 4 occurring during ambulation
or while transferring from chair or toilet. In com-
parison, the control group had a total of 12 falls: 4
from bed and 8 occurring elsewhere. There was no
significant difference in the number of bed falls
between the two groups (p = 1.00). None of the
falls in the study population resulted in physical
injury.

Although the use of mechanical restraints
and side rails was similar for both groups, in the
experimental group the utilization of mechanical
restraints to guard against bed falls declined. Me-
chanical restraints in this group were mostly em-
ployed to prevent patients from tampering with

TABLE 1

Falls Occurring in the Study and Nonstudy Population,
9-Month Evaluation Period of a Bed Alarm System

Bed Other

falls falls* Total

Experimental group

(n = 35)

Control group

(n = 35)

Nonstudy group

(n = 225)

Total

* During ambulation and transfers from a chair or toilet.

intravenous lines and urinary catheters and to
protect against falls when patients were out of
bed (sitting in a chair).

The experimental group utilized the system
for a total of 4,425 h, during which time a total of
143 alarms occurred. Of these, 120 alarms (84%)
were true and 23 alarms (16%) were categorized
as false. Malposition of the pressure sensor pad —
at the patient's shoulder level rather than under-
neath the buttocks — accounted for 14 (64%) of the
false alarms; reasons for the remaining 9 false
alarms (36%) are unclear.

Nurses were able to respond to 92% of the
alarm activations in less than 1 min (Table 2).
Eight alarms were answered either in 1-3 min or
in more than 3 min. The response time was not
indicated for 3 alarms.

Toileting or need to go to the bathroom was
the most common reason for transferring out of
bed and accounted for 78 (65%) of the true alarm
responses (Table 3). In 8 cases, the patient was
acutely ill or complained of pain, and in the re-
maining 34 instances, either patients were un-
able to communicate their reasons for getting up
from bed because of confusion or aphasia or the
reasons for alarm activation were unaccounted
for. Half of all patients attempting a bed transfer
after an alarm response were found to be either
sitting on the edge of the bed or climbing over the
side rail. In the remaining instances, patients
were found either standing by the bedside (20%),
in the bathroom (15%), or in the hallway (15%).

The alarm system was not associated with
any documented bodily harm to patients, nor did
the devices interfere with medical and nursing
care. Patients, family members, and nursing staff
universally approved the use of the system, and
several commented that using it was preferable to
using mechanical restraints.

Discussion

Bed alarm systems have been designed and
are utilized for one purpose: to prevent bed falls.
Toward this end, the effectiveness of a system can
be measured against several outcomes: the reli-
able operation of the devices in alerting nursing
staff when patients at risk of falling are partici-
pating in independent bed transfers; nurses' re-
sponse time in attending to the patient after an
alarm activation; and the extent to which use of
an alarm system decreases bed falls.

Operation of the Bed Alarms. In all cases,
the system functioned properly, activating an
alarm when pressure was removed from the sen-
sor pad. Nurses found the system to be user

Vol. 60 No. 6

PREVENTION OF FALLS— TIDEIKSAAR ET AL.

525

friendly, easy to install and operate and remain-
ing trouble free. The occurrence of false alarms
was the only factor that frustrated nurses and
inhibited use of the system. Staff removed the bed
alarm from one patient and replaced it with me-
chanical restraints. When false alarms were due
to the malpositioning of the sensor pad under-
neath the patient's shoulder level, the sensor pad
was repositioned underneath the patient's but-
tock to eliminate any further false alarms.

Nurse Response Time. Aside from one case,
nurses were able to respond to patients in a
timely fashion to avoid a fall, and they felt that
they had sufficient time to do so. In the one ex-
ception where a bed fall did occur, the nurse's
response time was recorded as greater than three
minutes. The circumstances surrounding this ep-
isode are unknown, but the event demonstrates
that these systems are only as effective in guard-
ing against bed falls as the nurse's response time.
The type of system used in this study may have
influenced nurse response times. The RN+ On-
Call bed monitoring system has a separate con-
sole located at the nurses' station for observing
alarms, as opposed to other models that operate
directly through the nursing call monitor. The
nurses may have been more attuned to respond to
alarm activations that originated from this sepa-
rate console and not from the nurse call system.

Bed Falls. The lack of statistical difference in
bed falls of system users and nonusers was unex-
pected but not totally surprising. During the
study period, both the experimental and control
groups sustained a low rate of bed falls. Conse-
quently, a failure to detect significance may be a
result of small sample size, rather than a function
of the system itself. Perhaps if a larger study pop-
ulation had been recruited or if a greater number
of bed falls had occurred, a positive association
between the use of the system and decreased bed
falls might have emerged.

An examination of fall occurrences in pa-
tients excluded from the study group because
they did not meet inclusion criteria lends some
support to this hypothesis. A total of seven falls
were experienced by this group, of which five
were bed falls (Table 1). A comparison of bed falls
in the experimental group (n = 1) and in the com-
bined control and nonstudy groups in = 9)
showed a greater propensity to fall from bed in
patients not using the system. Although the dif-
ference did not reach statistical significance (p -
0.3577), it does demonstrate an association be-
tween an increase in sample size and bed falls in
patients not using the system. Whether the pa-
tients in the nonstudy group were at increased

TABLE 2

Nurse Response Time to Alarm Activations, 9-Month
Evaluation Period of a Bed Alarm System

Number (%)

Less than 1 min

132 (92)

1-3 min

5(4)

More than 3 min

3(2)

Unknown*

3(2)

Total

143 (100)

* Response time not recorded.

risk for bed falls at the time, which might have
been prevented by a bed alarm, is unknown.

However, other explanations are possible.
First, the low incidence of bed falls in the study
population may reflect greater nursing and med-
ical effort to prevent falls in patients assumed to
be at high risk of falling. To prevent potential
falls, nurses hourly observed both groups of pa-
tients while patients were in bed and provided
assistance with transfers. There is the potential
for bias introduced by the nurses' awareness of
the intervention and perhaps by their anticipa-
tion of the goals of the study (the Hawthorne ef-
fect). The nurses' knowledge of patients being in-
cluded in the study may have caused them to
change their behavior toward the prevention of
bed falls. Also, physicians assigned to the unit are
geriatric fellows-in-training and geriatric attend-
ings who are more likely to search for modifiable
fall risk factors. A similar study on a general
medical unit, where nurses and physicians may
not be as focused on fall prevention, might yield a
greater sample size and significant findings. It
might also help to explain the low rate of bed falls
found in this study.

Second, the low rate of bed falls in the control
group may reflect an increased use of mechanical
restraints, applied either after the first fall or as
a precaution in patients observed to be participat-
ing in unassisted transfers. Anecdotal nursing re-
ports and retrospective chart reviews suggest
that possibility, but data are insufficient to truly
support it.

A third possibility is that the use of side rails
may have influenced bed fall rates. In both the
experimental and control groups, half side-rails
were universally employed as a fall preventive
measure. While previous research has noted an
association between the use of bed side-rails and
increased fall risk (3), in some instances side rails
may prevent falls. For example, we observed sev-
eral patients with poor bed mobility use the half
side-rail to safely transfer from bed, suggesting
that a partial side-rail may actually be used as a

526

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

TABLE 3

Reasons for True Alarms, 9-Month Evaluation of a Bed
Alarm System

Number {%)

Toileting

78 (65)

Acute illness

8(7)

Unknown

34 (28)

Total

120 (100)

falls preventive, a subject worthy of further re-
search.

Finally, the low fall rate may indicate that
patients with poor bed mobility either restrict
their number of bed transfers or call for assis-
tance through the nurse call system (available by
the bedside). However, this hypothesis is contra-
dicted in part by the large number of true alarm
system activations (n = 120) that occurred.
Whether the two groups of patients used the
nurse call system prior to exiting from the bed is
not known.

Efficacy of the Alarm System, Despite the
low rate of bed falls and lack of significant find-
ings, evidence supports the efficacy of the alarm
system. A comparison of users and nonusers dem-
onstrated a clinical trend toward decreased bed
falls. In addition, the 120 true alarm activations
that occurred in patients using the system repre-
sented potential bed falls that were avoided, with
the exception of one fall. Also, a bed alarm system
may protect against physical injury, because one
hip fracture occurred in the nonstudy group as a
result of a bed fall, and one patient in the control
group had a mechanical restraint complication;
the patient was found dangling by the bedside
with the vest restraint attached.

In addition to a reduction in bed falls, the
experimental group exhibited a clinical trend to-
ward decreased falls when out of bed. A compar-
ison of falls in the experimental and the control
group (Table 1) shows that patients using the sys-
tem had fewer falls (n = 4) while engaging in
out-of-bed activities than patients not on the sys-
tem (n - 8). Although the difference is not great,
it suggests that the system was protective in
alerting nurses that patients were getting out of
bed; it also allowed the nurses to provide assis-
tance for out-of-bed activities. This is supported
by the circumstances surrounding the falls in
each group. In the control group, the falls origi-
nated shortly after the patient left the bed. Con-
sequently, the nurses were not alerted ahead of
time when patients were engaging in out-of-bed
activities, and nurses were unable to offer assis-
tance. With the experimental group, the falls oc-

curred well after the patients had left the bed,
originating while walking or while transferring
from chair or toilet. These falls could not have
been avoided with the use of a bed alarm.

A tendency of the control and nonstudy
groups toward more falls raises the possibility
that observing patients hourly may not be helpful
in preventing falls. If future research supports
this assumption, then the use of nurses' time
might be better spent in performing other duties
important to patient care. This also suggests that
bed alarms may be a suitable alternative to fre-
quent patient observations by the staff and
thereby result in considerable cost savings for the
institution.

Potential Uses of Bed Alarms. Apart from
preventing falls, the findings suggest that bed
alarms may be beneficial in other ways. They
may be valuable in helping monitor clinical
change in patients. Eighty-three percent of the
true alarm responses were for either toileting
(frequent urination, nocturia) or acute illness
(congestive heart failure, chest pain), suggesting
that using a bed alarm to alert staff to changes in
the patient's medical condition may result not
only in more rapid interventions, but also in re-
duction of risk for falling. For example, nocturia
has been found to be a risk factor for falls (8) and
in some instances is an early indication of conges-
tive heart failure. If patients are treated at an
early stage, the frank onset of heart failure and
the likelihood of falling may be avoidable. Also,
attending to patients who require frequent toilet-
ing in a timely fashion may reduce both the risk
of falls and episodes of incontinence.

Summary

Bed alarms are an acceptable method of fall
prevention. The use of an alarm system for pa-
tients at risk of falling resulted in a clinical ten-
dency toward a decrease in falls from bed and
falls during out-of-bed activities. In addition,
such alarm systems may help reduce the use of
mechanical restraints and the frequency of nurse
monitoring to prevent bed falls. Bed alarms may
also be useful in monitoring clinical changes in
certain patients. The alarms were not associated
with any adverse patient effects, nor did they in-
terfere with patient care. The attitudes of pa-
tients, family members, and nurses toward the
utilization of an alarm system were positive.

Acknowledgments

The authors thank Jane Morris, R.N., Department of Nursing,
and the nursing staff of the Geriatric Evaluation and Treat-

Vol. 60 No. 6

PREVENTION OF FALLS— TIDEIKSAAR ET AL.

527

ment Unit, The Mount Sinai Medical Center, New York, New
York, for helping with data collection; and Tactilitics, Inc.,
Boulder, Colorado, for providing the RN + OnCall bed moni-
toring system.

References

1. Jones WJ, Smith A. Preventing hospital incidents: what

we can do. Nurs Manag 1989; 20(9):58-60.

2. Morgan VR, Mathison JH, Rice JC, et al. Hospital falls: a

persistent problem. Am J Public Health 1985; 75:775-
777.

3. Rubenstein HS, Miller FH, Postel S, et al. Standards of

medical care based on consensus rather than evidence:
The case of routine bed rail use for the elderly. Law Med
Healthcare 1983; ll(6):271-276.

4. Frengley JD, Mion LC. Incidence of physical restraints on

acute general medical wards. J Am Geriatr Soc 1986;
34:565-568.

5. Tinetti ME, Liu W, Marottoli RA, et al. Mechanical re-

straint use among residents of skilled nursing facilities.
JAMA 1991; 265(4):468--i71.

6. Tideiksaar R, Osterweil D. Prevention of bed falls: the

Sepulveda GRECC method. Geriatr Med Today 1989;
8:70-78.

7. Morton D. Five years of fewer falls. Am J Nurs 1989;

February:204-205.

8. Stewart RB, Moore MT, May FE, et al. Nocturia: a risk for

falls in the elderly. J Am Geriatr Soc 1992; 40:1217-
1220.

Geropsychology and
Neuropsychological Testing:

Role in Evaluation and Treatment of Patients with Dementia

Rajendra Jutagir, Ph.D.

Dementia is a condition of global cognitive dete-
rioration that is not due to clouding of conscious-
ness. A diagnosis is made by demonstrating im-
pairment in short- and long-term memory,
accompanied by impairment in at least one addi-
tional area of higher cortical function (language,
conceptualization), or personality change. Im-
pairment must be sufficient to interfere with
work or social activities. Progressive decline in
cognitive functions would suggest Alzheimer's
disease (1).

The Work Group established by the National
Institute of Neurological and Communicative
Disorders and Stroke and the Alzheimer's Dis-
ease and Related Disorders Association recom-
mended neuropsychological assessment to assist
in the diagnosis of Alzheimer's disease (2). Neuro-
psychological testing establishes whether the de-
gree of a patient's cognitive loss exceeds what one
would expect from normal aging. Apart from
identifying specific cognitive deficits, such testing
results in a profile of relative strengths and weak-
nesses. This can confirm whether deficits are
global or focal, and helps differentiate between a
variety of dementing disorders. In addition, test-
ing provides a baseline against which it is possi-
ble, by subsequent retesting, to document wheth-
er a patient's cognitive capacity has progressively
declined with time.

The psychologist working with older adults
encounters the same variety of mental disorders
as in younger adults, and utilizes similar (though

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, New York, NY. Address
reprint requests to Rajendra Jutagir, Ph.D., Box 1070, The
Mount Sinai Medical Center, One Gustave L. Levy Place, New
York, NY 10029.

528

sometimes modified) treatments (3). However, de-
mentia has a higher prevalence in the older
group, and is a source of such disability and an-
guish that it warrants special emphasis.

Although there is a tendency to neglect pa-
tients with dementia syndromes, who are often
seen as untreatable (4), a broad array of interven-
tions is available. Patients may be only mildly
impaired for several years, during which time
they can engage productively in high-level inter-
ventions such as psychotherapy. With modifica-
tion of technique, psychotherapy can be con-
ducted even into the moderate stages of dementia.
Apart from psychotherapy, there are numerous
psychosocial interventions. Treatment is also
available to caregivers of demented individuals;
by reducing stress and promoting understanding,
such treatment can considerably enhance the
quality of life of the patient.

The objectives here are to review the role of
neuropsychological testing in the evaluation of
dementia; and to examine psychotherapy and
psychosocial treatments for demented individuals
and their families, with emphasis on a psychoed-
ucational approach.

Neuropsychological Assessment

Disease processes in dementia produce
changes in the brain. A neuropsychological test
battery is administered to the patient to develop a
profile of cognitive functions, which is used to
make inferences about changes that have oc-
curred in brain structures.

A thorough test battery would assess the fol-
lowing cognitive functions: memory, attention,
language, conceptualization, visuospatial and
frontal systems functions. Recent memory is

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

GEROPSYCHOLOGY AND NEUROPSYCHOLOGICAL TESTING-JUTAGIR

529

tested by presenting the patient with verbal stim-
uli (stories, wordlists) or visual stimuli (designs,
objects) and having the patient recall them. Abil-
ity to learn new information is documented by
examining performance over repeated trials of a
memory task, which should show a learning
curve. After a delay of 20-30 minutes, the patient
is asked to recall stimuli learned earlier, which
would demonstrate whether there has been exces-
sive decay in the material learned. Remote mem-
ory can be tested by asking about verifiable his-
torical data (for example, the name of the
president during World War II). The modality of
test administration (verbal, visual) can be se-
lected to compensate for sensory impairment,
which is a common problem in the elderly. For
patients with speech deficits, recognition testing
and tests not requiring oral responses are used.
Methods of testing memory and other cognitive
functions in the elderly are reviewed in greater
depth elsewhere (5).

The Neuropsychological Profile and Its
Uses. The patient's neuropsychological profile
would be established by comparing performance
on these tests to established norms for the pa-
tient's age group, and charting areas of strength
and weakness. Even when test results fall within
the norms, performance may be impaired in rela-
tion to the level of the individual's own premorbid
functioning. Educational and occupational
achievement may assist in clarifying this situa-
tion.

This neuropsychological profile, consisting of
the level of performance in different areas of cog-
nitive functioning, is examined to arrive at a di-
agnosis. To characterize a dementia, global dete-
rioration would have to be documented by
impairment in memory and at least one other cog-
nitive ability. In Alzheimer's disease one would
further expect poor delayed recall, a strong indi-
cation of anterograde amnesia. Global impair-
ment with prominent frontal features, including
poor executive abilities and disinhibition, might
suggest Pick's disease (6). Yet another profile, one
that includes cognitive slowing and poor manip-
ulation of information, might suggest a subcorti-
cal dementing process (7). Often enough it is not
possible to clearly characterize the dementia.

Other disorders may be diagnosed from the
profile. One might suspect a focal lesion if, for
example, the patient is anomic but otherwise cog-
nitively intact, or if the primary findings are im-
pairment in attention and conceptualization. Pa-
tients with delirium would be expected to have
attentional deficits shown by impairment on
memory span and vigilance tasks. This is often

sufficiently pronounced to prompt discontinua-
tion of memory testing, since it calls into question
such patients' ability to register stimuli pre-
sented to them. Disorientation would also be ex-
pected in this disorder. In diagnosing pseudode-
mentia, qualitative aspects of testing play an
important role. One would look for inconsisten-
cies in testing, such as better performance on dif-
ficult than on easy items. Because mild depres-
sion is unlikely to manifest itself in significant
cognitive loss, mood would have to be quite de-
pressed for this diagnosis to be made.

Distinguishing whether change in cognitive
functioning is due to normal aging or to a disease
process is particularly difficult in early or mild
dementia, and screening instruments often lack
sensitivity in these cases. Thus patients who re-
peatedly complain of memory loss or other cogni-
tive difficulties may have scores within normal
limits on typical screening tests for dementia.
These patients should be diagnosed with in-depth
neuropsychological testing, which is probably the
most sensitive method of making this discrimina-
tion.

It is of course insufficient to know only that
the patient is demented. To clarify the etiology of
the dementia, the full array of causes must be
addressed. Medical examination to rule out infec-
tions, toxic or metabolic problems, and other neu-
rological involvement is a necessary part of the
diagnostic procedure. A brain scan is often help-
ful to rule out tumors, hydrocephalus, and stroke
activity. Once these possibilities are eliminated,
Alzheimer's disease and multi-infarct dementia
are the most likely causes of global cognitive im-
pairment in the elderly.

Interventions

The First Intervention: Review of Findings.

A critical moment in the care of the elderly de-
mented patient occurs immediately following
evaluation, when the provider and patient meet
to review the findings. Although rarely discussed
in the literature, this contact is of considerable
importance. It should be viewed as the first inter-
vention, because it lays the groundwork for later
interventions. The diagnosis is best presented by
a professional who has established a relationship
with the patient, to whom the patient can turn for
future care.

Since some causes of dementia are poten-
tially reversible, the case for early detection is
clear. However, there is some debate about the
wisdom of telling patients their diagnosis when
they have an irreversible dementia, Alzheimer's

530

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

disease in particular (8). Arguing against disclo-
sure is the uncertainty of the diagnosis, the per-
ceived paucity of therapeutic options, and the fear
that the stress of the diagnosis may exacerbate
the patient's symptoms. In favor of disclosure is
the right to self-determination, and the need for
full participation in formulating advance direc-
tives, financial planning, and other personal af-
fairs while the patient is still competent to ex-
press herself or himself. I take the position that
frankness in disclosing the diagnosis is generally
in the patient's best interest, promoting both self-
determination and confidence in the health-care
provider.

Clinical judgment on the patient's mental
status should determine how the diagnosis is pre-
sented. Many patients come to the evaluation al-
ready apprehensive of the possibility of Alzhei-
mer's disease. On those occasions when it is
confirmed, these patients may not be excessively
surprised. At the least, patients can tolerate
learning that they show some signs of cognitive
impairment which should be followed in the
months ahead to document whether it progresses.
This approach is helpful with very anxious or de-
pressed patients, with whom a treatment alliance
can be formed which will later allow deeper ex-
ploration of their symptoms and diagnosis. If the
patient is acutely emotionally disturbed, it may
be preferable to discontinue testing or withhold
the findings until the patient has been stabilized,
or to complete the evaluation on an inpatient ba-
sis.

Psychotherapy. Psychotherapy can help the
dementing patient with the immediate tasks of
adjusting to the diagnosis and becoming engaged
in planning for the future. Treatment necessarily
includes a psychoeducational component. Thus
the professional provides information about the
diagnosis and the disease, and makes recommen-
dations, while simultaneously responding to the
patient's emotional reactions and mobilization of
defenses. When the diagnosis is made early in the
dementing process, long-term psychotherapy is
also feasible for some patients. Deeper psychody-
namic exploration of the personality structure
and defenses can help the patient come to terms
with unresolved life conflicts.

Psychotherapy must be adapted to the stage
of the disease (9). With progression there is in-
creased emphasis on loss of function. Having
someone to discuss this with is reassuring to the
patient and can lead to better coping with deficits.
The therapist may have to be empathic in verbal-
izing the patient's feelings as self-expression di-
minishes. Still later, reality testing may become

compromised, with evidence of delusions or hal-
lucinations. It can reduce a patient's anxiety to
know that certain experiences are merely the
product of difficulty interpreting reality. For ex-
ample, one patient stated that he knew that the
bomb next to his bed was probably not real; in-
stead of becoming agitated and trying to flee the
bedroom, he turned over and went back to sleep.
Psychopharmacologic methods become increas-
ingly necessary in this stage of the disorder.

Psychoeducational Treatment for Family
Members. Some aspects of family psychoeduca-
tional treatment developed for mental illness (10)
can be adapted for use with the families of demen-
tia patients. Examining how family members re-
act to the patient's symptoms can help to manage
stresses of caregiving and minimize family bur-
den. The professional also serves an important
educational function. What is dementia? What
tests were done and what did they show? What
are the implications for the future? Will the pa-
tient have to go to a nursing home? Will the dis-
ease be inherited by family members? These
questions are commonly raised by families and
can be addressed in this context. Structured rec-
ommendations regarding future planning and use
of respite or home aides can also be made and
clarified. Providing continuity of care is critical so
that the family feels supported and knows where
to turn for help. Given the multiple needs of the
demented patient, the establishment of a treat-
ment team is helpful in ensuring that the patient
receives comprehensive care. Even in private
practice it is possible to establish relationships
with health-care providers in other disciplines
who, through consultations and referrals, can
serve as a team. Finally, connecting family mem-
bers with a support network, including groups
and social agencies, helps to reduce isolation and
the sense of enduring difficult times alone (11).

Conceptualizing dementia as a chronic men-
tal illness may put the disorder into perspective
for the health-care provider. In one view (12),
families must adapt to chronic mental illness in
three primary areas: interpersonal, instrumental,
and life course. Interpersonally, many feelings
are aroused between the patient and family mem-
bers. The latter may express anger toward the
patient because of his or her behavior, thinking
that the patient could perform normally by trying
harder. Their anger may be exacerbated by feel-
ing frustrated and let down by the health care
system that often fails to diagnose dementia ad-
equately, or to provide adequate support. The
therapist can address family anger by clarifying
that the patient is functioning as well as he or she

Vol. 60 No. 6

GEROPSYCHOLOGY AND NEUROPSYCHOLOGICAL TESTING-^UTAGIR

531

can, which may require providing concrete evi-
dence of organic dysfunction in the form of test
results. If the disease progresses, mourning the
loss of the person they knew becomes increasingly
important. Since patients often remain physically
healthy while their personality disintegrates un-
der the onslaught of the disease process, families
may not readily become aware of their sense of
loss and the emotions that accompany it. The
therapist can clarify the situation and help the
family accept this loss.

Instrumentally, the family has to adapt to
problems related to the patient's behavior. A ma-
jor source of burden to dementia caregivers is the
daily difficulty in functioning (losing things, in-
ability to get dressed), and problem behaviors
(wandering, combativeness). The therapist can
make practical suggestions about how to deal
with these, while at the same time exploring as-
pects of the caregiver's circumstances or person-
ality that make certain behaviors particularly
difficult to cope with.

Finally, life-course issues are raised because
patients can live for many years with a dementia.
A common pattern is for a close family member to
become completely consumed by the caregiving
task. With support and planning, the disease pro-
cess can become better integrated into family life
so that it does not demand the end of a normal
existence for this person. Family treatment can
explore family resources, build a sense of perspec-
tive with regard to the disease, and engender op-
timism in family members who may have felt
that they were unable to go on.

Summary

Cognitive loss in the aged should be evalu-
ated as early as possible, since some causes of
impairment may be reversible. Neuropsychologi-
cal testing is probably the most sensitive method
of detecting early or mild dementia. On comple-
tion of the evaluation, the first meeting with the
patient and family is critical in laying the
groundwork for subsequent interventions. Frank-
ness in discussing the diagnosis of dementia is
advocated, as it facilitates patient self-determina-
tion and enhances confidence in the health-care
provider. Interventions, including psychotherapy.

are presented in a psychoeducational framework,
where the aim is to provide information to the
patient and/or family while monitoring and re-
sponding to emotional reactions and defenses.
Conceptualizing dementia as a chronic mental ill-
ness may be helpful to the health-care provider.
Although there is no cure for Alzheimer's disease
or for multi-infarct dementia, treatments are
available in the psychological and psychosocial
domain. These have enormous potential for alle-
viating suffering and improving the quality of life
of patients and their families, and therefore
should not be overlooked.

References

1. American Psychiatric Association. Diagnostic and statis-

tical manual of mental disorders, 3rd ed.-rev. Washing-
ton, DC: American Psychiatric Association, 1987.

2. McKhann G, Drachman D, Folstein M, et al. Clinical di-

agnosis of Alzheimer's disease: report of the NINCDS-
ADRDA Work Group under the auspices of Department
of Health and Human Services Task Force on Alzhei-
mer's Disease. Neurology 1984; 34:939-944.

3. Myers WA, ed. New techniques in the psychotherapy of

older patients. Washington, DC: American Psychiatric
Press, 1991.

4. Cassel CK, Jameton AL. Dementia in the elderly: an anal-

ysis of medical responsibility. Ann Int Med 1981; 94:
802-807.

5. Albert M. Assessment of cognitive function in the elderly.

Psychosomatics 1984; 25:310-317.

6. Knopman DS, Christensen KJ, Schut LJ, Harbaugh RE,

Reeder T, Ngo T, Frey W. The spectrum of imaging and
neuropsychological findings in Pick's disease. Neurol-
ogy 1989; 39:362-386.

7. Albert ML. The 'subcortical dementia' of progressive su-

pranuclear palsy. J Neurol Neurosurg Psychiatry 1974;
37:121-130.

8. Drickhamer MA, Lachs MS. Should patients with Alzhei-

mer's disease be told their diagnosis? N Engl J Med
1992; 326:947-951.

9. Solomon K, Szwabo P. Psychotherapy for patients with

dementia. In: Morely J, Coe R, Strong R, Grossberg G,
eds. Memory function and aging-related disorders. New
York: Springer Publishing, 1992.

10. McFarlane WR. Family psychoeducational treatment. In:

Gurman AS, Kniskern DP, eds. Handbook of family
therapy, vol. 2. New York: Brunner/Mazel, 1991.

11. Zarit SH, Zarit JM. Families under stress: interventions

for caregivers of senile dementia patients. Psychother
Theory Res Practice 1982; 19:461-471.

12. Smyer MA, Birkel RC. Research focused on intervention

with families of the chronically mentally ill elderly. In:
Light E, Lebowitz B, eds. The elderly with chronic men-
tal illness. New York: Springer Publishing, 1991.

Clinical Evaluation of Dementia

Jonathan Koblenzer, M.D.

This article describes an approach to the clinical
diagnosis of the patient with possible dementia.
For both theoretical and practical reasons, these
patients pose special conceptual problems for the
clinician. For one thing, mental phenomena are
intangible and therefore hard to define and quan-
tify. Cardinal clinical phenomena such as mood
and affect, abstract thought and judgment — and
even more specific cortical functions such as ag-
nosia and the apraxias — do not have uniformly
accepted definitions. Clinicians can differ dra-
matically, depending on their education and spe-
cialty training, in how they define these phenom-
ena (1, pp. 391-405; 2, pp. 19-44). It is not
surprising that bedside clinical methods of deter-
mining a patient's mental state are similarly
variable (3), or that clinical syndromes can be de-
fined and ordered in different and overlapping
ways (1, pp. 103-107; 2, pp. 1-18).

Multiplicity is promoted because any mental
syndrome, including dementia, can be associated
with a great number of different diseases and
pathologic processes. Dementia is a clinical diag-
nosis based on finding impairment in a number of
functions of the mind. These impairments can
only be found in a clinical interview. Abnormal
blood tests and scans do not make the diagnosis of
dementia. There are no biologic markers diagnos-
tic of the dementia syndrome; conversely, there
are no mental symptoms pathognomonic of a par-
ticular dementing disease process in the brain.
More practically, our knowledge of brain func-
tioning in relation to mental events in health and
disease is growing explosively, so that all current
knowledge has a provisional quality.

Two strategies have been developed over the

From the Department of Psychiatry, The Mount Sinai Medical
Center, New York, New York. Address reprint requests to
Jonathan Koblenzer, M.D., Department of Psychiatry, Box
1230, The Mount Sinai Medical Center, One Gustave L. Levy
Place, New York, NY 10029.

last 30 years to address these issues. The first is
the development of a standardized descriptive
language of psychiatry, as well as operational-
ized, criterion-based psychiatric diagnosis as em-
bodied in the Diagnostic and Statistical Manual of
Mental Disorders, of which the third revised edi-
tion (DSM-III-R) is the latest (1, pp. 103-107).
Thus the diagnostic inclusion and exclusion cri-
teria that define a disorder are explicitly listed.
The modified Jones criteria for acute rheumatic
fever are an example from another field of medi-
cine (4).

The second strategy is the development of
standardized questionnaires, such as the Mini-
Mental State Exam (MMSE) (5) or the Hamilton
Depression Scale (6), that quantify the intensity
of mental symptoms. Of note, instruments like
the MMSE are best used to quantify symptoms in
previously diagnosed patients and do not replace
diagnostic tools such as DSM-III-R.

Definitions

Several definitions of the dementia syndrome
have been published and are used clinically (1,
pp. 103-107; 2, pp. 1-18). An ideal definition
would combine maximum diagnostic sensitivity
and specificity with ease of use. Unfortunately, no
such definition has been developed. The DSM-
III-R criteria (1, pp. 103-107) (ruled box 1),
though not universally accepted (2, pp. 1-18), are
easy to use and have been shown to be reliable (7).

The cardinal symptoms of dementia are
short- and long-term memory loss. By definition,
memory is always impaired in dementia. Short-
term memory is tested by asking the patient to
repeat three unrelated words five minutes after
being presented with them. Long-term memory
can be more difficult to test; the results often re-
quire corroboration from the family or friends of
the patient. This is particularly important in Alz-
heimer's disease (AD), where patients can main-

532

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

CLINICAL EVALUATION OF DEMENTIA— KOBLENZER

533

1. Diagnostic Criteria for Dementia
(Diagnostic and Statistical Manual of
Mental Disorders 3rd rev. ed.)

A. Demonstrable evidence of
impairment in short-
and long-term memory.

B. At least one of the following:

(1) impairment in abstract
thinking

(2) impaired judgment

(3) other disturbances of higher
cortical function, such as
aphasia, apraxia, agnosia, and
"constructional difficulty"
(constructional apraxia)

(4) personality change (i.e.,
alteration or accentuation of
premorbid traits)

C. The disturbance in A and B
significantly interferes with work
or usual social functioning.

D. Not occurring exclusively during
the course of delirium.

E. Either (1) or (2):

(1) There is evidence from history,
physical examination,
laboratory tests of a specific
organic factor judged to be
etiologically related to the
disturbance.

(2) In the absence of such evidence,
an etiological organic factor can
be presumed if the disturbance
cannot be accounted for by any
nonorganic mental disorder
(e.g., major depression
accounting for cognitive
impairment).

Reprinted with permission from DSM-III-R, p. 107.

tain the appearance of good social functioning un-
til quite late in their illness.

By definition, memory disturbance must be
accompanied by at least one other type of cogni-
tive impairment; often it is associated with sev-
eral. Abstraction, which can be defined as think-
ing in terms of categories rather than specifics
(8), can be tested by eliciting interpretations of
proverbs; responses may be abstract or concrete.
A concrete interpretation of the proverb "Don't
cry over spilt milk," for example, is that milk is
easily cleaned up or that more can be bought at
the store. A more abstract response is that past
events are irrevocable, cannot be influenced by
present wishes, and are therefore not worth wor-
rying about.

Judgments are global evaluations of a situa-
tion, with a view to possible actions (9). Judgment
is best evaluated by assessing the patient's ability
to perform day-to-day responsibilities, informa-
tion usually readily available from the history.
Actions such as giving money away to strangers
or disrobing in public are not uncommon exam-
ples of impaired judgment in patients with de-
mentia.

Aphasias, constructional and ideomotor
apraxias, and agnosia are symptoms that attend
damage to specific cortical areas. The best way to
define and test some of these functions remains
controversial (2, pp. 19-44; 3). Broadly speaking,
these symptoms can be evaluated thus: aphasia,
by asking a patient to repeat a sentence; construc-
tional apraxia, asking for a copy of a figure; ideo-
motor apraxia, asking a patient to follow a com-
mand; and agnosia, asking patient to recognize
and name common objects. These tests are all
part of the MMSE.

Personality comprises the stable and predict-
able patterns of reaction to and experience of life
events. Personality traits are usually established
by adolescence; change in personality is an im-
portant sign that can be associated with demen-
tia.

DSM-III-R criteria require that any cognitive
deficits significantly interfere with social and vo-
cational functioning. Again, this is usually evi-
dent from a well-taken history. In addition, vari-
ous functional scales, such as the Instrumental
Activities of Daily Living Scale (lADLS) (10), can
help quantify the degree of overall functional im-
pairment. Of note, there is evidence of a correla-
tion between degree of cognitive impairment on
the MMSE and performance in lADLS (11).

Ruling out delirium is crucially important.
Delirium is the "great mimic" of psychiatry and,
by definition, has an often reversible organic

cause. A delirious patient can have a wide variety
of psychiatric syndromes, including dementia,
but delirium is usually not present in patients
who are able to attend (1, pp. 100-103), which can
be tested by asking the patient to immediately
repeat a series of digits presented at one-second
intervals. Any patient who can repeat five digits
or more is determined to be attentive (2, pp. 21—
24) and delirium is not likely. Inattention, de-
fined as inability to repeat five digits, has a wide
differential diagnosis, including delirium, mania,
depression, anxiety, and psychosis. Making a syn-
dromal diagnosis in the elderly inattentive pa-
tient can be extremely difficult.

534

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Finally, there must be at least presumptive
evidence of an organic etiology of the dementia
syndrome (ruled box 2).

A great virtue of these criteria is that a di-
agnosis can usually be made quickly at the bed-
side. Properly used, instruments such as the
MMSE and the Global Deterioration Scale (GDS)
(12) can help make the diagnosis of dementia and
at the same time quantify its severity. Folstein et
al. found that MMSE scores of 23 or less were

predictive of the presence of a dementing process
(5) and that the scores varied with the severity of
functional impairment. Modified criteria using
DSM-III, the predecessor to DSM-III-R, have had
a positive predictive value of 82%-90% in pa-
tients with Alzheimer's disease (13). On the other
hand, these criteria are insensitive for clinical
signs of early dementia (14). The results of
screening tests, such as the MMSE and the GDS,
must be interpreted with caution. Sensory defi-
cits, such as hearing and visual loss, aphasias,
and educational level, can all confound the inter-
pretation of results.

Etiology

There are a number of causes of the dementia
syndrome (ruled box 2), but in Western countries
most cases are associated with either Alzheimer's
disease (22%-70%) or multi-infarct dementia
(MID) (8%-47%) (2, pp. 5-8; 15); other cases are
caused by a variety of conditions. The marked
variability is most likely due to selection bias in
the studies, differences in diagnostic criteria, and
the practical difficulty in making an absolutely
definitive diagnosis of dementia. In the following
section, common or otherwise important (e.g., re-
versible) etiologies and their presentations are
discussed.

Alzheimer's Disease. AD is characterized by
a progressive, steady decline in cognitive func-
tioning beginning in mid-to-late life. Memory
dysfunction and word-finding problems are early
symptoms, along with poor judgment and disori-
entation to place. Impaired performance of de-
layed visual and verbal recall tasks is a sensitive
sign of early dementia (16). As the disease pro-
gresses, fluent aphasia supervenes, with worsen-
ing memory and disorientation. Psychiatric man-
ifestations, such as depression, psychosis, and
personality change (usually irritability), as well
as agitation and wandering behaviors, may ap-
pear in as many as 90% of AD patients (17). Late
AD is characterized by severely impaired cogni-
tion, mutism, incontinence, and agitation. Con-
tractures, seizures, infections, and coma super-
vene in the terminal phase. The prevalence of AD
is highly correlated with age, varying from a rate
of 4.4/100,000 in people under 60 to 1431.7/
100,000 in those over 80 (18).

The diagnosis of AD is made by excluding
other possible causes of dementia (1, pp. 119-121)
(see below). In the living patient, it is not feasi-
ble at present to exclude all other etiologies of
dementia, so there is usually a degree of un-
certainty in the clinical diagnosis. One set of
diagnostic criteria, developed by the National

2. Classification of Some Common Causes
of Dementia

Degenerative dementias

Alzheimer's disease
Pick's disease
Parkinson's disease
Huntington's disease

Vascular dementias

Viral dementias
HIV-1 encephalopathy
Progressive multifocal
leukoencephalopathy
Jakob-Creutzfeld disease

Bacterial dementias

Neurosyphilis
Lyme disease

Metabolic dementias

Vitamin deficiency

Bi2, folate

Thiamine

Niacin
Endocrinopathy

Hypothyroidism

Hypoglycemia

Hyperparathyroidism
Heavy metal intoxication
Alcohol and drug abuse
Prescription and nonprescription drugs

(e.g., psychotropics)
Uremia
Liver failure

Chronic obstructive pulmonary disease
Congestive heart failure

Hydrocephalus

Trauma

Space-occupying lesions

Subdural hematomas
Tumors

Adapted from ref. 38.

Vol. 60 No. ()

CLINICAL EVALUATION OF DEMENTIA— KOBLENZER

535

Institute of Neurological and Communicative
Disorders and Stroke and the Alzheimer's Dis-
ease and Related Disorders Association Work
Group (19), reflects this by labeling patients as
having "possible," "probable," or "definite" AD,
depending on the available clinical evidence.
Characteristic neuropathologic findings on biopsy
or autopsy material are required for the latter
diagnosis.

Multi-Infarct Dementia. In contrast to the
steady and progressive onset of Alzheimer's dis-
ease, MID is characterized by abrupt and step-
wise appearance of the symptoms of the dementia
syndrome (1, pp. 121-123), along with evidence of
stroke or lateralizing neurologic signs and a
patchy distribution of deficit. Some mental func-
tions are preserved quite well, whereas others are
significantly impaired. The presence of conditions
that are risk factors for cerebrovascular disease,
such as atrial fibrillation, diabetes mellitus, or
hypertension, supports the diagnosis. Evidence of
infarcts on computed tomography or magnetic
resonance imaging is confirmatory. MID com-
prises a heterogeneous group of conditions, in-
cluding large- and small-vessel thrombotic and
embolic disease (2, pp. 170-171). Also in contrast
to Alzheimer's disease, MID occurs in younger pa-
tients and affects men more than women (20).

Alzheimer's Disease and Multi-Infarct De-
mentia. One autopsy study showed that MID and
AD coexisted in about 20% of dementia patients
(21), not an unexpected finding in view of the fre-
quency of each of these conditions. Mixed demen-
tia of this type is suspected when a patient evi-
dences features of each. Common scenarios
include the patient with progressive onset of de-
mentia and a few lacunar infarcts on imaging
tests, or a patient with a cognitive impairment
that seems disproportionate to the size and loca-
tion of an infarct.

Parkinson's Disease. A mild-to-moderate de-
mentia can occur in 15%-20% of patients with
Parkinson's disease (22), characterized by impov-
erishment and slowing of the cognitive processes,
concreteness, and indecisiveness; predominantly
cortical functions, such as speech, are usually
spared. This combination of symptoms has been
described as a subcortical dementia (23). It is im-
portant to rule out depression, which can occur in
40%— 60% of patients with Parkinson's disease
(24) and can mimic a dementing process
("pseudodementia") .

"Reversible" Dementias. Among the demen-
tias accounted as reversible are the more common
dementing processes for which some evidence ex-
ists that the initiation of a specific treatment will

halt or even reverse the progression of the dis-
ease, making the diagnosis of these conditions
particularly important. On the other hand, it is
debatable how reversible some of these dementias
truly are (25), since the number of cases studied is
relatively small and there are often confounding
variables in the case reports. In addition, some
treatments are invasive and attended by some
risk of complications.

Vitamin B^g deficiency can be associated
with dementia and depression in the absence of
anemia (26). The dementia is often associated
with subacute combined degeneration of the spi-
nal cord. Treatment with parenteral vitamin Bj^2
can induce more or less complete recovery occur-
ring in days to weeks. Rarely, folate deficiency
produces a similar clinical picture (27).

Slowed cognition is not uncommon in pa-
tients who have significant hypothyroidism. More
rarely, the clinical picture can progress to frank
dementia or delirium with delusions and halluci-
nations (28). Partial or complete reversal of these
syndromes is possible with thyroid replacement.

Neurosyphilis was formerly a common cause
of institutionalization for dementia and psycho-
sis, usually 20 years after the initial infection.
Delusions of grandeur associated with dementia,
affective instability, and hallucinations were all
part of the clinical picture. In the antibiotic era
this classic picture is rare, because many patients
are exposed to antibiotics during the course of
their illness (29). Instead, patients may offer a
nonspecific picture of dementia associated with
gradual loss of vision, loss of ankle jerks, and ab-
normalities of the pupillary responses, among
other signs.

Normal-pressure hydrocephalus, first de-
scribed in 1965 (30), is suspected in patients with
the clinical triad of dementia, gait disturbance
(apraxic or "magnetic" gait), and incontinence,
and in whom neuroimaging reveals lateral ven-
tricles enlarged out of proportion to sulcal size.
Ventricular shunts can dramatically relieve
symptoms in these patients. Serial lumbar punc-
tures have been used to select patients for this
procedure (31).

Chronic subdural hematoma can attain large
size and remain asymptomatic; not uncommonly,
it is associated with the dementia syndrome (32).
Delirium and focal neurologic findings can both
supervene.

Miscellaneous. Pick's disease, Huntington's
disease, and Jakob-Creutzfeld disease are three
uncommon dementing illnesses that are notewor-
thy because of their often striking clinical pic-
tures. It is beyond the scope of this article to dis-

536

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

CUSS them, but any clinician evaluating patients
with dementia needs to be familiar with their di-
agnosis.

Differential Diagnosis

Despite the unambiguous diagnostic criteria
for dementia, it can be hard to distinguish pa-
tients with dementia from those with delirium,
those with the primary "functional" disorders of
depression or psychosis, and those with focal cor-
tical deficits, such as aphasia or amnesia.

Delirium is common in elderly patients, par-
ticularly those with dementia; in addition, delir-
ium can itself mimic dementia almost exactly. As
mentioned above, the presence of distractability
(1, pp. 100-103), or inattention, can distinguish
between the two entities. In the inattentive pa-
tient who appears demented, every effort to iden-
tify the common causes of delirium must be made
because treatments are usually available. In el-
derly patients, these causes are most commonly
drug toxicities, infections, and metabolic and
electrolyte disturbances (33). Administration of
all unnecessary medications should be discontin-
ued, and doses of necessary drugs should be re-
duced to a minimum. Often this is sufficient to
clarify the picture, but sometimes only after sev-
eral weeks of drug washout.

If there is no evidence of other likely causes
of delirium, an electroencephalogram (EEG) is
sometimes helpful. The diffuse slowing of the pre-
dominant rhythms that often occurs in delirium
is not as common in mild dementia or in func-
tional illness in the elderly. If the EEG is normal,
the inattention may be due to psychiatric syn-
dromes, such as anxiety, depression, mania, hal-
lucinations, or formal thought disorder. These
syndromes may occur independently of the de-
menting process or be a consequence of it. In these
cases, consultation with a geriatric psychiatrist
can be helpful in clarifying the diagnosis.

In addition to causing inattention in the pa-
tient with dementia, depression and psychosis
can, like delirium, mimic dementia itself. Some
depressed patients can, in all respects, look de-
mented (34); with treatment this picture remits.
Clinical criteria for distinguishing "pseudode-
mentia" of depression from dementia proper have
been proposed (34). However, the positive and
negative predictive values of these criteria are
not known. The recent onset of neurovegetative
signs of depression, such as sleep and appetite
disturbance or psychomotor agitation, may be
helpful in making the diagnosis. Often the clini-
cal distinction is difficult, and consultation with a

geriatric psychiatrist can help clarify the diagno-
sis. Recent work suggests that "pseudodementia"
may be a herald of dementia proper (35).

Similarly, some patients with schizophrenic
illnesses may appear to be demented, especially if
they evidence extreme social withdrawal, mut-
ism, catatonia, or a severe formal thought disor-
der. Again, psychiatric consultation may be help-
ful.

Aphasia and amnesia may be components of
the dementia syndrome but are not in themselves
sufficient to make the diagnosis. Patients with
these conditions are often incorrectly diagnosed,
either because they are severely disoriented due
to amnesia or because they have difficulty com-
municating due to their aphasia. A careful exam-
ination can usually demonstrate that they do not
suffer from the more global impairments of de-
mentia.

Diagnosis and Evaluation

The basic strategy in the evaluation and di-
agnosis of the patient with apparent dementia is
to accomplish the following two goals. The first is
to identify the psychopathologic features of the
patient's illness in order to make a syndromal di-
agnosis. This depends on a good clinical inter-
view, to distinguish dementia from delirium, de-
pression, or other "functional" or organic
psychiatric syndromes (as above). The second goal
is to identify the etiology of the dementia syn-
drome. Here the history, physical examination,
and selected laboratory tests and special investi-
gations can be essential. The main reason for
making an etiologic diagnosis is to identify either
treatable or heredofamilial dementias. In either
case, the information obtained can be useful to
the patient and his or her family. For example, up
to 20% of patients with Alzheimer's disease are
reported to have inherited their illness with an
autosomal dominant pattern (36).

All patients with suspected dementia need a
complete history and physical examination, as
the possible causes of dementia are numerous
(ruled box 2). The history should focus on the on-
set of the illness, drug and alcohol exposure, nu-
tritional history, exposure to toxins, heavy met-
als, or infectious agents, medical illnesses and
medication use (including nonprescription medi-
cations), and family history of dementia or psy-
chiatric illness. A physical examination, with
careful attention to the nervous system, can con-
firm or reveal the presence of serious systemic or
neurologic illness that may explain part or all of
the etiology of the dementia.

Vol. 60 No. 6

CLINICAL EVALUATION OF DEMENTIA— KOBLENZER

537

Laboratory studies should include at least a
complete blood cell count, complete chemical
screen with liver function tests, urinalysis, mea-
sures of vitamin B12 and folate levels, serum drug
screen, rapid plasma reagin, thyroid function
tests, electrocardiogram, chest x-ray, and comput-
erized tomography of the brain (37). This infor-
mation should suffice to diagnose the vast major-
ity of dementias; further investigation is usually
unnecessary. Little current evidence supports
routine comprehensive neuropsychologic testing
in the differential diagnosis of the dementia syn-
drome (2, p. 40). Such testing may be diagnosti-
cally helpful in patients who have early dementia
or atypical clinical features.

Depending on the clinical situation, an elec-
troencephalogram, lumbar puncture, serum
heavy metal screens, test for human immunode-
ficiency virus, and brain biopsy may be helpful, as
well as neurologic and psychiatric consultation. If
studies for dementia reveal no etiologic cause, the
patient most likely has Alzheimer's disease; in-
farcts suggest multi-infarct dementia or mixed
(MID and AD) dementia, and other positive find-
ings suggest a specific etiology or mixed dementia
(with Alzheimer's disease), as discussed above.

Summary

The diagnosis of dementia is a clinical one
that can only be made with the rigorous use of
clearly defined diagnostic criteria in an interview
setting. The diagnosis cannot be made by labora-
tory tests or imaging techniques alone. In the el-
derly, a number of other psychiatric syndromes
must be distinguished from dementia, including
delirium, psychosis, depression, and mania.
Sometimes it is not possible to make this distinc-
tion with confidence.

Most dementias are caused by Alzheimer's
disease or multi-infarct dementia, and the proba-
ble etiologic diagnosis can be made fairly quickly.
However, a number of etiologies are possible, and
some patients require extensive study. Some un-
certainty is usual in the diagnosis of the living
patient because routine brain biopsy is not feasi-
ble. Geropsychiatric and neurologic consultants
can be helpful in selected cases.

Acknowledgments

I thank Jeffrey Newcorn, M.D., and Paul Aisen, M.D., for their
careful reading and criticism of the manuscript.

References

1. American Psychiatric Association. Diagnostic and statis-
tical manual of mental disorders, 3rd rev. ed. Washing-
ton, DC: American Psychiatric Press, 1987.

2. Cummings JL, Benson DP. Dementia: a clinical approach,

2nd ed. Boston: Butterworth-Heinemann, 1992.

3. Benton A. Visuoperceptual, visuospatial, and visuocon-

structive disorders. In: Heilman KM, Valenstein E, eds.
Clinical neuropsychology. 2nd ed. New York: Oxford
University Press, 1985:151-179.

4. Stollerman G. Rheumatic fever. In: Wilson JD, Braun-

wald E, Isselbacher KJ, et al. Harrison's principles of
internal medicine, 12th ed. New York: McGraw-Hill,
1991:935.

5. Folstein MF, Folstein SE, McHugh PR. "Mini-Mental

State": a practical method for grading the cognitive
state of patients for the clinician. J Psychiatr Res 1975;
12:189-198.

6. Hamilton M. A rating scale for depression. J Neurol Neu-

rosci Psychiatry 1960; 23:56-62.

7. Forette F, Henry JF, Orgogozo JM, et al. eds. Reliability

of clinical criteria for the diagnosis of dementia: a lon-
gitudinal multicenter study. Arch Neurol 1989; 46:646—
648.

8. Frosch J. Psychodynamic psychiatry: theory and practice.

Madison, CT: International Universities Press, 1990:
401.

9. Sims A. Symptoms in the mind. London: Bailliere Tindal,

1988:112-113.

10. Lawton MP, Brody EM. Assessment of older people: self-

maintaining and instrumental activities of daily living.
Gerontologist 1969; 9:179-186.

11. Warren EJ, Grek A, Conn D, et al. A correlation between

cognitive performance and daily functioning in elderly
people. J Geriatr Psychiatry Neurol 1989; 2:96-100.

12. Reisberg B, Ferris SH, DeLeon MJ, Crook T. The Global

Deterioration Scale for assessment of primary degener-
ative dementia. Am J Psychiatry 1982; 139:1136-1139.

13. Mendez MF, Mastri AR, Sung JH, Frey WH II. Clinically

diagnosed Alzheimer's disease: neuropathological find-
ings in 650 cases. Alzheim Dis Assoc Disord 1992; 6:
35^3.

14. Rosenman S. The validity of the diagnosis of mild demen-

tia. Psychol Med 1991; 21:923-934.

15. Skoog I, Nilsson L, Palmertz B, Andreasson L-A, Svan-

borg A. A population-based study of dementia in 85-
year-olds. N Engl J Med 1993; 328:153-158.

16. Sahakian BJ, Morris RG, Evenden JL, Heald A, Philpot

M, Robbins TW. A comparative study of visuospatial
memory and learning in Alzheimer-type dementia and
Parkinson's disease. Brain 1988; 111:695-718.

17. Mendez MF, Martin RJ, Smyth KA, Whitehouse PJ. Psy-

chiatric symptoms associated with Alzheimer's disease.
J Neuropsychiatry Clin Neurosci 1990; 2:28-33.

18. Schoenberg BS, Kokmen E, Okazaki H. Alzheimer's dis-

ease and other dementing illnesses in a defined United
States population: incidence rates and clinical features.
Ann Neurol 1987; 22:724-729.

19. McKhann G, Drachman D, Folstein M, Katzman R, Price

D, Stadlan E. Clinical diagnosis of Alzheimer's disease:
report of the NINCDS-ADRDA Work Group under the
auspices of the Department of Health and Human Ser-
vices Task Force on Alzheimer's Disease. Neurology
1984; 939-944.

20. Morimatsu M, Hirai S, Muramatsua A, Yoshikawa M.

Senile degenerative brain lesions and dementia. J Am
Geriatr Soc 1975; 23:390^06.

21. Jellinger K. Neuropathological aspects of dementias re-

sulting from abnormal blood and cerebrospinal fluid dy-
namics. Acta Neurol Belg 1976; 76:83-102.

538

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

22. Brown RG, Mardsden CD. How common is dementia in

Parkinson's disease? Lancet 1984; 1:1262-1265.

23. Cummings JL. Subcortical dementia: neuropsychology,

neuropsychiatry, and pathophysiology. Br J Psychiatry
1986; 149:682-697.

24. Mayeux R, Stern Y, Williams JBW, et al. Clinical and

biochemical features of depression in Parkinson's dis-
ease. Am J Psychiatry 1986; 143:756-759.

25. Friedland RP. "Normal"-pressure hydrocephalus and the

saga of the treatable dementias. JAMA 1989; 262:
2577-2581.

26. Pallis CA, Lewis PD. The neurology of gastrointestinal

disease. Philadelphia: WB Saunders, 1974.

27. Fehling C, Jagerstad M, Lindstrand K, Elmqvist D. Fo-

late deficiency and neurological disease. Arch Neurol
1974; 30:263-265.

28. Swanson JW, Kelly JJ Jr, McConahey WM. Neurological

aspects of thyroid dysfunction. Mayo Clin Proc 1981;
56:504-512.

29. Lukehart SA, Holmes KK. Syphilis. In: Wilson JD,

Braunwald E, Isselbacher KJ, et al. Harrison's princi-
ples of internal medicine, 12th ed. New York: McGraw-
Hill, 1991:654-655.

30. Adams RD, Fisher CM, Hakim S, Ojemann RG, Sweet

WH. Symptomatic occult hydrocephalus with "normal"

cerebrospinal fluid pressure: a treatable syndrome. N
Engl J Med 1965; 273:117-126.

31. Haan J, Thommer RTWM. Predictive value of temporary

external lumbar drainage in normal pressure hydro-
cephalus. Neurosurgery 1988; 22:388-391.

32. Perlmutter I, Gobies C. Subdural hematoma in older pa-

tients. JAMA 1961; 176:212-214.

33. Lipowski Z. Delirium: acute confusional states. New

York: Oxford University Press, 1990:109-140.

34. Wells CE. Pseudodementia. Am J Psychiatry 1979; 136:

895-900.

35. Alexopoulos G, Young RC, Dreyer BS, et al. Late-onset

depression. Psychiatr Clin North Am 1988; 4:109-116.

36. Zubenko GS, Huff J, Beyer J, Auerbach J, Teply I. Famil-

ial risk of dementia associated with a biological subtype
of Alzheimer's disease. Arch Gen Psychiatry 1988; 45:
889-893.

37. Brown MM, Hachinski VC. Acute confusional states, am-

nesia, and dementia. In: Wilson JD, Braunwald E, Is-
selbacher KJ, et al, eds. Harrison's principles of inter-
nal medicine, 12th ed. New York: McGraw-Hill, 1991:
192.

38. Whitehouse PJ, ed. Dementia. Philadelphia: FA Davis,

1993.

Osteoporosis in the Aged

Carolyn Murray, M.D., and Diane E. Meier, M.D.

Osteoporosis occurs when bone mineral density
decreases to the point that bone fractures in the
presence of minimal or no trauma. Clearly, pre-
venting bone loss or attaining optimal peak bone
mass as a young adult can minimize susceptibil-
ity to this disorder. Early research in osteoporosis
emphasized prevention of the rapid bone loss that
occurs in the early postmenopausal years. Yet the
effects of osteoporosis are felt much later in life.
This paper summarizes the evaluation and treat-
ment of elderly women with osteoporosis, draw-
ing on the minimal research on osteoporosis in
the elderly.

Epidemiology

Approximately 250,000 hip fractures and
500,000 vertebral fractures occur each year in the
United States. The lifetime risk of a hip fracture
for a 50-year-old white woman is 15.6%, with a
median age for hip fracture of 79 years (1). The
lifetime risk of Colles' fracture is 15.0% at 50
years of age, with a median age of 66 years for the
fracture. Approximately 32% of 50-year-old white
women will experience a vertebral fracture, al-
though Obrant et al. found that only half of all
vertebral fractures come to medical attention (2).
Twenty-seven percent of 65-year-old and 18.3% of
80-year-old white women can expect to suffer a
first vertebral fracture (1). Thus, by and large,
elderly women have the disease.

A recent prospective study showed that a de-
crease in bone mass by two standard deviations
from that of a normal 35-year-old conferred a
four-to-sixfold increase in risk for vertebral frac-
ture, whereas a single prior vertebral fracture in-

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, One Gustave L. Levy Place,
New York, New York. Address reprint requests to: Diane E.
Meier, M.D., The Mount Sinai Medical Center, Box 1070, One
Gustave L. Levy Place, New York, NY 10029.

creases risk of recurrence by four-to-fivefold.
Compared to women with the highest bone den-
sity and no vertebral fractures, women with the
lowest bone mass and one fracture were 25 times
more likely to fracture during each year of follow-
up (3).

Bone Metabolism in
Elderly Women

Bone loss rates in elderly women are less
than half those in perimenopausal and early post-
menopausal women, 1%-1.5% as compared to
2%-3% per year. Although the annual bone loss
rate in older women is lower, Nordin et al. esti-
mated that the majority of bone loss between 55
and 75 years of age was due to aging (62%) rather
than to menopause (38%) (4). Although some pre-
vious studies of bone loss in the elderly showed a
slowing of bone loss with age, two recent cross-
sectional studies found that bone loss continued
into the ninth decade of life. The Framingham
study identified a linear decline in bone mass in
the proximal femur and proximal radius, but not
the ultradistal radius (5). Annual changes in bone
mineral density were approximately - 0.53% for
the trochanter, - 0.68% for the femoral neck, and
-0.88% for the proximal radius, but only
- 0.16% for the ultradistal radius. This pattern of
loss continued through the late 70s and 80s and
into the mid-90s. Steiger et al. found that declines
in the lumbar spine bone density were consider-
ably less than the hip or radial losses (6). Bone
density declined 16% in the spine, 18%-20% in
the radius, and 26% in the femoral neck between
65 and 85 years of age. A linear relationship be-
tween bone loss and age was again identified into
the mid-80s. Since bone loss continues into the
ninth decade, the effect of late postmenopausal
bone loss can be devastating.

Many clinicians assume that osteoporosis in
the elderly is untreatable and that future frac-

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

539

540

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

tures are inevitable. This nihilistic approach is
not justified. Although the bone density of most
elderly women decreases below the theoretical
fracture threshold, the inverse relationship be-
tween bone mineral density and fractures contin-
ues below the fracture threshold; that is, the
higher the bone density, the lower the fracture
risk (1, 3). The fracture threshold does not repre-
sent an all-or-nothing phenomenon. For example,
clinicians would consider a patient with a blood
pressure of 210/110 mm Hg and a patient with a
blood pressure of 165/95 mm Hg as hypertensive,
and even a 10% decrease in blood pressure would
be considered beneficial to both patients, al-
though the first patient would remain hyperten-
sive. Similarly, individuals below the theoretical
fracture threshold may, in fact, receive signifi-
cant benefit from a slight decrease in their risk of
fracture through preventing further bone loss or
providing a small increase in bone mass from ac-
tive treatment.

Bone histomorphometry in the aged has
shown both increased and decreased bone turn-
over in different individuals. A decrease in the
connectivity of trabecular plates has been noted.
Furthermore, new bone formation declines with
decreased repair of microfractures, increasing the
risk of fracture (7).

Risk Factors for Fracture in
The Elderly

Bone density clearly decreases with age. In
fact, virtually all older women are below the theo-
retical fracture threshold and are at increased
risk for fracture. Nonetheless, continued bone
density loss only increases the already high risk
for fracture. Numerous studies have shown in-
creased risk of fracture with each incremental de-
cline in bone mineral density (1, 3). Paganini-Hill
et al. identified multiple risk factors for hip frac-
ture, including the presence of diabetes, thinness,
tallness, cigarette smoking, alcohol intake, inac-
tivity, early menopause, and decreased sunlight
exposure, many of which are clearly associated
with decreased bone mineral density (8).

The elderly are also at increased risk of fall-
ing. One-third of persons over 65 fall each year;
however, only 6% of these falls lead to fractures.
The increased fall risk in the elderly may be re-
lated to comorbid illnesses, polypharmacy, or
other concomitants of aging, such as postural hy-
potension, Parkinson's disease, stroke, or lower
extremity disability such as arthritis. Visual loss
from cataracts, glaucoma, or macular degenera-
tion also places elderly individuals at risk for

falls. Medications that increase fall risk include
antihypertensive drugs, sedatives, antipsychotic
agents, and tricyclic antidepressants (9).

Therapy of the Older
Osteoporotic Woman

There is now definitive proof of beneficial
treatment in elderly women with osteoporosis;
therefore, each individual patient should be eval-
uated for therapy that may provide the most ben-
efit and acceptable risk. An individual with low
bone density alone may be treated less aggres-
sively than a patient who has experienced multi-
ple vertebral fractures. Three modalities of treat-
ment for osteoporosis are discussed below:
calcium, vitamin D, and antiresorptive agents
such as estrogen.

Calcium. Calcium deficiency is common as
women age, because of both poor gastrointestinal
absorption and low calcium intake. Absorption of
calcium from the intestine decreases 20%-25%
between ages 40 and 60 owing to the combined
effects of loss of estrogen at menopause and aging
(10). In the 1980 NHANES survey, the median
daily intake of calcium among women older than
44 years was 475 mg, less than half the recom-
mended calcium intake for postmenopausal
women (11). The roles of low-calcium diets and
poor calcium absorption in the elderly osteoporot-
ic are not well understood, but there is some evi-
dence that a relative calcium insufficiency may
stimulate bone resorption by increasing levels of
parathyroid hormone (12).

Recent studies have confirmed earlier find-
ings that calcium stabilizes bone loss more effec-
tively in late postmenopausal women than in
early menopausal women. A study by Dawson-
Hughes et al. showed calcium supplementation
with 500 mg of calcium citrate malate stabilized
spinal bone loss and actually increased bone den-
sity in the radius and femoral neck among late
menopausal women with a dietary calcium intake
of 400 mg or less (13). Reid et al. recently showed
a beneficial effect of 1000 mg of calcium supple-
mentation among women 10 years past meno-
pause who had a relatively high baseline dietary
intake of calcium of 750 mg (14). The rate of bone
loss was decreased at the hip, and bone density
actually increased in the lumbar spine. A 14-year
prospective study found that a dietary calcium
intake greater than 765 mg/day was associated
with a 60% reduction in hip fracture risk (15).
Although no fracture data showing benefit from
calcium supplementation alone are available, a
recent study of 3,270 elderly French women liv-

Vol. 60 No. 6

OSTEOPOROSIS IN THE AGED— MURRAY AND MEIER

541

ing in nursing homes showed fewer nonvertebral
fractures among women randomly assigned to re-
ceive 1,200 mg of calcium and 800 lU of vitamin
D than among those receiving placebo. Women
receiving the combination of calcium and vitamin
D experienced 27% fewer hip fractures and 26%
fewer nonvertebral fractures than women on pla-
cebo (16).

The metabolic effects of six months of cal-
cium supplementation include significant de-
clines in both serum parathyroid levels and
1,25-OH vitamin D levels (17). Bone turnover pa-
rameters, including serum bone Gla protein and
urinary hydroxy proline excretion levels, did not
change. The mechanism of action for calcium sup-
plementation in late postmenopausal women may
be related to suppression of age-related increases
in parathyroid hormone levels that might in-
crease bone resorption.

Finally, Orwoll et al. conducted a histomor-
phometric study of the effects of 1,200 mg of cal-
cium carbonate on bone in elderly osteoporotic
women with an average dietary calcium intake of
854 mg (7). The baseline bone biopsy revealed de-
pressed bone remodeling rates without evidence
of osteomalacia. Repeat bone biopsies showed an
unexpected decrease in both mineralization lag
time and the duration of remodeling cycles, sug-
gesting that a relative calcium deficiency in late
postmenopausal osteoporosis may impair the
mineralization phase of remodeling.

In summary, recent data suggest that cal-
cium supplementation with 500-1,000 mg/day
stabilizes both trabecular and cortical bone loss in
women during the late postmenopausal years.
This effect appears to be most powerful in women
with low dietary calcium intakes of less than 400
mg/day, but it remains beneficial for women with
higher dietary calcium intakes.

Achlorhydria, cost, and compliance compli-
cate calcium supplementation in the elderly.
Achlorhydria increases considerably with age, de-
creasing the level of free acid in the unfed stom-
ach; one study showed that almost 40% of women
over age 50 have no free acid in the unfed state
(18). Calcium carbonate is not absorbed well
without gastric acid secretion and, therefore,
should not be given on an empty stomach to el-
derly patients. Absorption studies have found
that calcium citrate is better absorbed through
the gastrointestinal tract than calcium carbon-
ate, especially when taken without food (19). Ab-
sorption of calcium from the gastrointestinal
tract is also dose dependent, with minimal in-
creases in calcium absorption in doses greater
than 500 mg (19).

Calcium carbonate remains the least expen-
sive, most widely available source of calcium sup-
plementation, which must be considered when
prescribing to the elderly patient. If preferred for
these reasons, calcium carbonate should be given
with meals, and all calcium supplements should
be prescribed in divided doses if the patient is
taking more than 500 mg of calcium.

Compliance with calcium supplements is re-
lated to their side effects, especially constipation
in the elderly. Starting with low doses of calcium
supplementation and building up to the desired
dose may be helpful; however, many elderly pa-
tients will need to change from calcium carbonate
to a less constipating source of calcium, such as
calcium citrate, calcium citrate malate, or a liq-
uid form, calcium glubionate. Unfortunately, cal-
cium citrate malate has been commercially avail-
able only in fortified fruit juices (20).

The overall required dose of calcium supple-
mentation varies with the patient's dietary cal-
cium intake. Although recent studies suggest a
total intake of 800 mg may be sufficient, a more
conservative guideline of approximately 1,200 mg
of total calcium intake may compensate for indi-
vidual variation in absorption. This total intake
of calcium is unlikely to cause clinical problems
unless the patient is hypercalcemic or hypercal-
ciuric. Hypercalciuria may be suggested by a per-
sonal or family history of kidney stones or an un-
expectedly low bone density (more than 2 SD
below normal for age), which may be due to renal
calcium wasting.

Vitamin D. Vitamin D insufficiency has been
well described in the elderly population. Mc-
Kenna suggests that 25% of community-dwelling
elderly people have low vitamin D values in the
winter (21). Estimates of vitamin D intake by
community-dwelling older people is approxi-
mately 100 lU per day — half the recommended
daily allowance, and one-quarter of the dose
available in most multivitamins (11). A recent
study of 22 homebound elderly patients aged 68-
96 found one-third had serum levels of vitamin D
less than 25 mmol/L, despite a dietary intake of
467 lU daily (22). These authors suggest that 400
lU of dietary intake is not sufficient in home-
bound elderly patients without sun exposure. In
addition to decreased oral intake of vitamin D in
the elderly, aging appears to decrease skin and
renal production of vitamin D.

Vitamin D insufficiency can affect bone loss
in the elderly in at least two ways: by decreasing
calcium absorption and by leading to osteomala-
cia. Osteomalacia is a bone disorder caused by
vitamin D deficiency where normal bone matrix

542

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

is formed but is not mineralized. The prevalence
of low-grade osteomalacia among patients with
intracapsular hip fracture in the United States
was reported as 25% in 32 patients studied by
Sokoloff (23). Furthermore, 40% of 172 patients
hospitalized with hip fracture had low 25-OH vi-
tamin D levels, reflecting decreased vitamin D
intake (24). Vitamin D insufficiency also in-
creases secretion of parathyroid hormone, which
in turn increases resorption of bone (11). Dawson-
Hughes et al. randomized 249 postmenopausal
women with usual daily vitamin D intakes of 100
lU to receive either 400 lU of vitamin D or pla-
cebo, in addition to 377 mg/day of calcium (25).
Women taking vitamin D decreased their winter-
time spinal bone loss significantly (-0.54%, in-
stead of - 1.22% taking placebo). There was a sig-
nificant one-year increase in bone density of
0.88% in the group taking vitamin D and calcium,
compared to an increase of 0.15% in those taking
calcium and placebo.

The French nursing home study of over 3,000
ambulatory women randomized to receive either
1.2 g of calcium and 800 lU of vitamin D or pla-
cebo found over a 25% decline in both hip and
nonvertebral fractures in the treatment group. At
baseline, this population had low serum vitamin
D values and slightly elevated parathyroid hor-
mone values, which normalized in the group on
calcium and vitamin D supplementation (16).

Overall, these studies support the recommen-
dation that elderly individuals supplement their
low dietary intake of vitamin D with 400 lU of
vitamin D available in a multivitamin. Home-
bound elderly persons deprived of sun exposure
should increase their vitamin D intake to 800 lU/
day.

The data on pharmacologic doses of vitamin
D or calcitriol (l,25-OH2 vitamin D) in the treat-
ment of osteoporosis are contradictory. Some
studies found higher fracture rates or greater loss
of bone density on calcitriol, but others showed
decreased bone loss or decreased fractures (26-
30). The most recent study utilizing low doses of
calcitriol (0.25 \xg twice a day) found markedly
fewer vertebral fractures among women receiving
calcitriol for three years than among women re-
ceiving calcium alone (28). Elevations of serum
calcium or creatinine were infrequent in the low
doses used in this study; however, increased se-
rum and urinary calcium levels were identified in
previous studies (29, 30). Thus, patients treated
with calcitriol must calcium-restrict their diets to
avoid symptomatic hypercalcemia. At present,
calcitriol is not indicated for the treatment of pri-
mary osteoporosis outside of a research setting.

Antiresorptive Therapy in the Elderly. Es-
trogen replacement therapy (ERT) remains the
treatment of choice in the prevention of osteopo-
rosis and in the treatment of women with osteo-
porosis under age 70. If estrogen replacement
therapy is continued, its efficacy is sustained at
least until age 70. However, it is not clear that
newly prescribing ERT to older women will be
beneficial. Most studies of ERT have focused on
women younger than 65.

Two randomized clinical trials including
women over 65 showed a significant increase in
bone density for women receiving estrogen. One
trial of 39 women showed a significant decrease
in vertebral fracture rates in the group receiving
transdermal estrogen; however, few women over
age 70 were included in this study (31). A popu-
lation-based cohort study in Sweden found that
women who had ever used estrogen had a lower
relative risk for hip fracture of 0.79 (0.68-0.93)
(32). However, patients older than 60 had no sig-
nificant decrease in hip fracture, whether they
had ever used estrogen in the past or had recently
used estrogens in the last five years. A recent
case-control study found a protective effect for
women taking estrogen, the relative risk for
hip fracture being 0.55 (0.31-0.85) (33). Estrogen
appeared to be less effective for women over
age 80.

Use of ERT in women over 70 years of age
must be based on studies in younger women.
Therefore, when prescribing hormone replace-
ment therapy, it is essential to assess individual
risks for fracture, cardiovascular disease, and
breast cancer. A recent meta-analysis of hormone
replacement therapy suggests an overall benefit
with increased life expectancy for women at high
risk of hip fracture (1.1 years) or heart disease
(1.6 years) (34). Women at high risk for breast
cancer may increase their lifetime probability of
breast cancer as a result of ERT.

Although studies of estrogen and cardiovas-
cular disease have shown benefit from estrogen
replacement therapy, including a decreased rela-
tive risk in both coronary artery disease and mor-
tality, benefit from ERT for women over 70 has
not been consistently shown. For example, the
ten-year prospective observational study of
nurses 30-63 years old at entry found a relative
risk of mortality of 0.56 among current estrogen
users. The relative risk in the oldest age group,
however, suggested a trend toward current use
increasing relative risk of mortality to 1.35 (not
significant) (35). Other studies following patients
into their 70s showed decreases in age-adjusted
mortality rates. The Framingham study found a

Vol. 60 No. 6

OSTEOPOROSIS IN THE AGED— MURRAY AND MEIER

543

protective effect for younger women, but, again, a
nonsignificant adverse effect was observed in
older women (36).

Although controversy exists over the role of
ERT in breast cancer, a recent meta-analysis sug-
gests a 30% increase in risk after 15 years of ERT
(37). Estrogen replacement therapy should only
be considered for women willing and able to un-
dergo appropriate gynecologic follow-up, regular
breast examination, and mammography.

Calcitonin. Calcitonin, a hormone derived
from the thyroid gland, has been shown to de-
crease bone loss, but no prospective data are
available to show a protective effect against frac-
tures. A recent case-control study of 2,086 women
with hip fractures and 3,532 age-matched con-
trols found a protective effect for women taking
calcitonin and calcium with a relative risk of hip
fractures of 0.63 (confidence interval, 0.44-0.90)
(33). The relative risk was 0.82 (0.63-1.07) for
calcium alone and 0.78 (0.48-1.27) for calcitonin
alone.

Calcitonin acts as an analgesic for some pa-
tients with acute compression fractures, although
many patients do not tolerate the side effects of
nausea and flushing or refuse parenteral admin-
istration. Furthermore, calcitonin therapy is
very expensive; moreover, antibodies to the hor-
mone frequently develop after several years of
use. Calcitonin in the form of a nasal spray is
widely utilized throughout the European conti-
nent. This preparation of the agent is better tol-
erated than parenteral injection, yet the drug re-
mains very expensive, and its efficacy in fracture
prevention is not clearly demonstrated.

Bisphosphonates. Bisphosphonates, new
medications for osteoporosis, are undergoing ex-
tensive clinical trials for use in both the treat-
ment and prevention of osteoporosis. Bisphospho-
nates act by inhibiting resorption of bone by
the osteoclast. Etidronate, a bisphosphonate ap-
proved for treatment of Paget's disease, decreased
bone loss and vertebral fractures in two small
studies of osteoporotic women (38, 39). Unfortu-
nately, the two studies of the effects of eti-
dronate on vertebral compression fractures were
complicated by either unusually high or low frac-
ture rates in the control groups, obfuscating eval-
uation of treatment effect.

Several clinical trials are now under way in
this country evaluating the effect of newer
bisphosphonate medications on bone loss and pre-
vention of vertebral fractures. Initial studies in
Europe have shown that some of the experimen-
tal bisphosphonates prevent postmenopausal bone
loss (40).

Thiazide Diuretics. Thiazide diuretics de-
crease urinary calcium excretion and may de-
crease the risk of osteoporotic fractures by im-
proving net calcium balance. Seven of eight
studies reviewed suggest protection from hip frac-
ture, higher bone density, or slower bone loss in
thiazide users (41-48). No randomized clinical
trials of thiazides for osteoporosis have been un-
dertaken.

Who to Treat

No guidelines have been established for the
treatment of osteoporosis in elderly women. Re-
cent studies suggest that most elderly women
should receive appropriate calcium supplementa-
tion to approximate a daily intake of 1,000-1,200
mg of calcium. Most individuals should also re-
ceive 400 lU of vitamin D, especially during the
winter months in northern latitudes. Increased
doses of vitamin D should be considered for the
institutionalized elderly and for individuals with
malnutrition, malabsorption, or evidence of osteo-
malacia.

Since most women fall below the fracture
threshold as they age, studies to determine sec-
ondary causes of osteoporosis are recommended
for those who have lower bone density than ex-
pected for their age, a rapid succession of frac-
tures, or other clues that postmenopausal osteo-
porosis is not the sole perpetrator of thin bones.
As examples, a rapid succession of vertebral frac-
tures may be due to multiple myeloma, elevated
serum calcium levels may be due to hyperpar-
athyroidism, and a family history of kidney
stones may suggest renal hypercalciuria.

Studies for secondary causes must be tailored
to the individual patient, but might include an
immunoprotein electrophoresis, 24-hr urine test
for calcium, creatinine, Bence Jones proteins or
free Cortisol, 25-hydroxy vitamin D levels, sensi-
tive thyroid-stimulating hormone, parameters of
malabsorption such as carotene, iron, or vitamin

levels, or parathyroid hormone measures. In
elderly patients, a measure of intact parathyroid
hormone may be more accurate than the midmol-
ecule assay, especially when renal insufficiency is
present.

If a secondary cause of bone loss is present,
appropriate therapy should be instituted, such as
thiazide diuretics when the patient has renal hy-
percalciuria.

If patients have lower than expected bone
density or begin fracturing, antiresorptive ther-
apy should be considered. Antiresorptive agents
are most effective in high bone turnover disease.

544

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Measures of bone turnover, such as osteocalcin or
pyridinoline crosslinks, may help in deciding
whom to treat. The choice of antiresorptive agent
remains a complex decision in light of the lack of
data in the elderly. The cardiovascular benefits of
estrogen may not be evident over age 70, the in-
jection of calcitonin may be difficult because of
visual changes, arthritis, or compliance with in-
jections, and the gastrointestinal side effects and
poor absorption of bisphosphonates may limit
their acceptability. Once again, choice of therapy
must be tailored to each individual to maximize
compliance and safety and reduce bone loss.

References

1. Cummings SR, Black DM, Rubin SM. Lifetime risks of

hip, CoUes' or vertebral fracture and coronary heart
disease among white postmenopausal women. Arch In-
tern Med 1989; 149:2445-2448.

2. Obrant KJ, Bengner U, Johnell O, Nilsson BE. Increasing

age-adjusted risk of fragility fractures: a sign of in-
creasing osteoporosis in successive generations? Calcif
Tissue Int 1989; 44:157-167.

3. Ross PD, Davis JW, Epstein RS, Wasnich RD. Pre-exist-

ing fractures and bone mass predict vertebral fracture
incidence in women. Ann Intern Med 1991; 114:919-
923.

4. Nordin BEC, Need AG, Chatterton BE, Horowitz M, Mor-

ris HA. The relative contribution of age and years since
menopause to postmenopausal bone loss. J Clin Endo-
crinol Metab 1990; 70:83-88.

5. Hannan MT, Felson DJ, Anderson JJ. Bone mineral den-

sity in elderly men and women: results from the
Framingham osteoporosis study. J Bone Min Res 1992;
7:547-553.

6. Steiger P, Cummings SR, Black DM, Spencer NE, Genant

HK. Age-related decrements in bone mineral density in
women over 65. J Bone Min Res 1992; 7:625-632.

7. Orwoll ES, McClung MR, Oviatt SK, Recker RR, Weigel

RM. Histomorphometric effects of calcium and calcium
plus 25 hydroxy-vitamin D3 therapy in senile osteopo-
rosis. J Bone Min Res 1989; 4:81-88.

8. Paganini-Hill A, Ross RK, Gerkins VR, Henderson BE,

Arthur M, Mack TM. Menopausal estrogen therapy and
hip fractures. Ann Intern Med 1981; 95:28-31.

9. Ray WA, Griffin MR, Schaffner W, Baugh DK, Melton U.

Psychotropic drug use and the risk of hip fracture. N
Engl J Med 1987; 316:363-369.

10. Heaney RP, Recker RR, Stegman MR, Moy AJ. Calcium

absorption in women: relationship to calcium intake,
estrogen status and age. J Bone Min Res 1989; 4:469-
475.

11. Krall EA, Sahyoun N, Tannenbaum S, Dallal GE, Daw-

son-Hughes B. Effect of vitamin D intake on seasonal
variations in parathyroid hormone secretion in post-
menopausal women. N Engl J Med 1989; 321:1777-
1783.

12. Lips P, Wiersinga A, Van Ginkel FC, et al. The effect of

vitamin D supplementation on vitamin D status and
parathyroid function in elderly subjects. J Clin Endo-
crinol Metab 1988; 67:644-649.

13. Dawson-Hughes B, Dallal GE, Krall EA, Sadowski L,

Sahyoun N, Tannenbaum. A controlled trial of the ef-

fect of calcium supplementation on bone density in post-
menopausal women. N Engl J Med 1990; 323:878-883.

14. Reid IR, Ames RW, Evans MC, Gamble GD, Sharpe SJ.

Effect of calcium supplementation on bone loss in post-
menopausal women. N Engl J Med 1993; 328:460^64.

15. Holbrook TL, Barrett-Connor E, Wingard DC. Dietary

calcium and risk of hip fracture: a 14 year prospective
population study. Lancet 1988; 2:1046-1049.

16. Chapuy MC, Arlot ME, Duboeuf F, et al. Vitamin D3 and

calcium to prevent hip fractures in elderly women. N
Engl J Med 1992; 327:1637-1642.

17. Kochersberger G, Westlund R, Lyles KW. The metabolic

effects of calcium supplementation in the elderly. J Am
Geriatr Soc 1991; 39:192-196.

18. Recker RR. Calcium absorption and achlorhydria. N Engl

J Med 1985; 313:70-73.

19. Harvey JA, Zobitz MM, Pak CYC. Dose dependency of

calcium absorption: a comparison of calcium carbonate
and calcium citrate. J Bone Min Res 1988; 3:253-257.

20. Heaney RP. Calcium supplements: practical consider-

ations. Osteo Int 1991; 1:65-71.

21. McKenna MJ. Differences in vitamin D status between

countries in young adults and the elderly. Am J Med
1992; 93:69-77.

22. Gloth FM, Tobin JD, Sherman SS, Hollis BW. Is the rec-

ommendation for vitamin D too low for the homebound
elderly? J Am Geriatr Soc 1991; 39:137-141.

23. SokolofTL. Occult osteomalacia in American patients with

fracture of the hip. Am J Surg Pathol 1978; 2:21-30.

24. Doppelt SH, Neer RM, Daley M, Bourret C, Schiller A,

Holick MF. Vitamin D deficiency and osteomalacia in
patients with hip fractures. Orthop Trans 1983; 7:512-
513.

25. Dawson-Hughes B, Dallal GE, Krall EA, Harris S, Sokoll

LJ, Falconer G. Effect of vitamin D supplementation on
wintertime and overall bone loss in healthy postmeno-
pausal women. Ann Intern Med 1991; 115:505-512.

26. Ott SM, Chestnut CH. Calcitriol treatment is not effective

in postmenopausal osteoporosis. Ann Intern Med 1989;
110:267-274.

27. Jensen GF, Christiansen C, Transbol I. Treatment of post-

menopausal osteoporosis. A controlled therapeutic trial
comparing oestrogen/gestagen, 1,25-dihydroxy-vitamin
Dg and calcium. Clin Endocrinol 1982; 16:515-524.

28. Tilyard MW, Spears GFS, Thomson J, Dovey S. Treat-

ment of postmenopausal osteoporosis with calcitriol or
calcium. N Engl J Med 1992; 326:357-362.

29. Gallagher JC, Goldgar D. Treatment of postmenopausal

osteoporosis with high doses of synthetic calcitriol. Ann
Intern Med 1990; 113:649-655.

30. Aloia JF, Vaswani A, Yeh JK, Ellis K, Yasumura S, Cohn

SH. Calcitriol in the treatment of postmenopausal os-
teoporosis. Am J Med 1988; 84:401^08.

31. Lufkin EG, Wahner HW, O'Fallon NM, et al. Treatment

of postmenopausal osteoporosis with transdermal estro-
gen. Ann Intern Med 1992; 117:1-9.

32. Naessen T, Persson I, Adami HO, Bergstrom R, Bergkvist

L. Hormone replacement therapy and the risk for first
hip fracture: a prospective, population based cohort
study. Ann Intern Med 1990; 113:95-103.

33. Kanis JA, Johnell O, Gullberg B, et al. Evidence for effi-

cacy of drugs affecting bone metabolism in preventing
hip fractures. Br Med J 1992; 305:1124-1128.

34. Grady D, Rubin SM, Petitti DB, et al. Hormone therapy to

prevent disease and prolong life in postmenopausal
women. Ann Intern Med 1992; 117:1016-1037.

35. Stampfer MJ, Colditz GA, Willett WC, et al. Postmeno-

Vol. 60 No. 6

OSTEOPOROSIS IN THE AGED— MURRAY AND MEIER

545

pausal estrogen therapy and cardiovascular disease:
ten year follow-up from the Nurses' Health Study. N
Engl J Med 1991; 325:756-762.

36. Eaker ED, Castelli WP. Coronary heart disease and its

risk factors among women in the Framingham study.
In: Eaker E, Packard B, Wenger NK, Clarkson TB, Ty-
roler HA, editors. Coronary heart disease in women.
New York: Haymarket Doyina, 1987:122-130.

37. Steinberg KK, Thacker SB, Smith SJ, et al. A meta-anal-

ysis of the effect of estrogen replacement therapy on the
risk of breast cancer. JAMA 1991; 265:1985-1990.

38. Watts NB, Harris ST, Genant HK, et al. Intermittent cy-

clical etidronate treatment of postmenopausal osteopo-
rosis. N Engl J Med 1990; 323:73-79.

39. Storm T, Thamsborg G, Steiniche T, Genant HK, Son-

ensen OH. Effect of intermittent cyclical etidronate
therapy on bone mass and fracture rate in women with
postmenopausal osteoporosis. N Engl J Med 1990; 322:
1265-1271.

40. Reginster JY, Deroisy R, Denis D, et al. Prevention of

postmenopausal bone loss by tiludronate. Lancet 1989;
2:1469-1471.

41. Wasnich RD, Benfante RJ, Yano K, Heilbrun L, Vogel

JM. Thiazide effect on the mineral content of bone. N
Engl J Med 1983; 309:344-347.

42. Ray WA, Downey W, Griffin MR, Melton LJ. Long-term

use of thiazide diuretics and risk of hip fracture. Lancet
1989; 1:687-690.

43. LaCroix AZ, Wienpahl J, White LR, et al. Thiazide di-

uretic agents and the incidence of hip fracture. N Engl
J Med 1990; 322:286-290.

44. Heidrich FE, Stergachis A, Gross KM. Diuretic drug use

and the risk of hip fracture. Ann Intern Med 1991; 115:
1-6.

45. Felson DT, Sloutkis D, Anderson JJ, Anthony JM, Kiel

DP. Thiazide diuretics and the risk of hip fracture: re-
sults from the Framingham study. JAMA 1991; 265:
370-373.

46. Transbol I, Christensen MS, Jensen GF, Christiansen C,

McNair P. Thiazide for postponement of postmeno-
pausal bone loss. Metabolism 1982; 31:383-386.

47. Adland-Davenport P, Mckenzie MW, Notelovitz M, Mc-

Kenzie LC, Prendergast JF. Thiazide diuretics and
bone mineral content in postmenopausal women. Am J
Obstet Gynecol 1985; 152:630-634.

48. Wasnich RD, Ross PD, Heilbrun LK, Vogel JM, Yano K,

Benfante RJ. Differential effects of thiazide and estro-
gen upon bone mineral content and fracture prevalence.
Obstet Gynecol 1986; 67:457-462.

The Principle of Autonomy and the

Older Patient

Diane E. Meier, M.D.

The rising interest in the ethical aspects of
modern medical practice results from two recent
and historically unprecedented phenomena: the
growth in numbers of dependent elderly persons,
and the dizzying velocity of advances in medical
technology that have occurred over the last sev-
eral decades. The dramatic 28-year gain in life
expectancy over the first 50 years of the twentieth
century is equivalent to that gained in the 4,000
years preceding this century, from the Bronze
Age (2000 Bc) to the year 1900.

There are several reasons for this remark-
able increase in length of life, and they are sim-
pler than one might guess; the extension of the
life span is primarily due to better public and pre-
ventive health measures, such as clean water, ba-
sic nutrition, public sanitation measures, and
routine prenatal care. This improvement in life
expectancy predated the advent of antibiotics and
the high technologic diagnostic and therapeutic
procedures that are currently breaking our
health care budget. The recent resurgence of tu-
berculosis despite the availability of effective
drug treatment is eloquent testimony to the pri-
mary contribution of social and public health
factors, primarily poverty, on the incidence of
disease.

As a consequence of this recent gain in life
expectancy, the numbers of older adults in the
United States have reached new highs: 12% of the

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, New York, New York. Ad-
dress reprint requests to Diane E. Meier, M.D., Department of
Geriatrics and Adult Development, The Mount Sinai Medical
Center, One Gustave L. Levy Place, Box 1070, New York, NY
10029.

population is now 65 years old or older, a percent-
age predicted to increase to 22% by the year 2040.
One often-quoted statistic is that the fastest-
growing segment of the population is the group of
persons over age 85 (1). The impact of this growth
in number of older adults will be felt throughout
society, but nowhere more powerfully than on the
practice of medicine.

The second factor behind rising interest in
medical ethics is the influence of the so-called
technology imperative on the modern practice of
medicine, whereby the existence of a technology
with any possibility of benefit, no matter how re-
mote, seems to mandate its use (2). In the early
years of this century, medical decisions for ill and
elderly persons were greatly simplified by the
physician's inability to substantially alter the
course of diseases (usually infectious and quick)
leading to death. Because little power to affect the
ultimate course of illness was available to either
doctors or patients, conflict over decisional au-
thority and appropriate applications of technol-
ogy did not exist.

Changes in these aspects of medical practice
have resulted from exponential technical ad-
vances in ability to prolong life despite chronic
irreversible illnesses, such as end-stage lung,
heart, and kidney disorders, and debilitating neu-
rologic degenerative diseases, such as stroke and
Alzheimer's disease. This very recent shift, from
death and dying as an unchangeable and usually
short-term process to a state of life amenable to
multiple interventions and prolongation over a
long period of time, has fueled interest in the
study of medical ethics and an attempt to apply
ethical principles to medical care and medical de-
cisions for persons approaching the end of their
lives.

546

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

AUTONOMY AND THE OLDER PATIENT— MEIER

547

Autonomy and Frail
Elderly Patients

In part because of the increasing power of
medicine to prolong life and the uncertain benefit
of such a technologic imperative for some older
adults, a growing movement for patients' self-de-
termination, based on the ethical principle of in-
dividual self-determination, has begun to have a
major influence on health-care policy. The in-
forming value of this patients' rights movement is
the principle of autonomy that embodies respect
for persons and their free choice. Well grounded
in the legal precedent of this country, it was best
expressed by Justice Benjamin Cardozo in a 1914
decision: "Every human being of adult years and
sound mind has a right to determine what shall
be done with his own body" (3). Recent court de-
cisions, such as the Quinlan and Cruzan cases,
have also consistently upheld the primacy of in-
dividual free will in medical decision making,
particularly in considerations of termination of
life support (4, 5).

This concern for the promotion of patient au-
tonomy in medical decision making has also
found expression in advance directives and living
wills, whereby people can state their wishes re-
garding those life-sustaining treatments that
they may or may not wish to have applied to them
at the end of their life. Similarly, the recent avail-
ability of proxy appointments, by which people of
intact mental capacity may appoint another indi-
vidual to make medical decisions on their behalf
in case of any future mental incapacity, is an-
other method of trying to advance autonomy in
health-care decisions. These efforts have made
their way into law in the form of the 1991 federal
Patient Self-Determination Act, which requires
that all institutions receiving Medicare or Medic-
aid reimbursement provide adult patients, on ad-
mission, with written information on advance di-
rectives and on the right of adults to refuse any
life-sustaining treatments (6). In New York
State, and in many other states as well, the law
permits patients to appoint proxies for health-
care decisions in case any future mental incapac-
ity develops (7). These laws were enacted to en-
hance the power of individuals if and when they
find themselves literally at the mercy of modern
medicine; the laws are an attempt at maximizing
the chances that any treatments employed are
truly in accord with the patient's values and
preferences.

This seems all to the good. But the applica-
tion of these strict autonomy-based principles to
medical decisions in frail older persons is fraught
with contradiction and ambiguity. Many features

of being an older adult necessarily restrict the
unfettered primacy of patient self-determination:
the progressive series of personal and functional
losses that tend to accompany aging; the often
unconscious ageism of a society that places great
emphasis on rugged self-reliance and indepen-
dence, characteristics that are often lost in older
adults because of the burdens of chronic illnesses
and associated disability that eventually afflict
most of us if we are fortunate enough to reach
old age; the associated consequence of increasing
dependency on other people for daily living, a re-
ality that is often profoundly distressing both to
older adults and to their families and that may
make true autonomous decision making nearly
impossible; and finally, the spiritual and emo-
tional recognition of the relative imminence of
death that accompanies this stage of human de-
velopment. These aspects of being old often create
a special vulnerability in the older patient that
requires appropriate sensitivity and awareness
on the part of the physician and that prohibits a
facile reliance on patient autonomy as a basis for
medical decision making.

Problems in Applying
The Principle of Autonomy to
Frail Patients

What are the drawbacks to utilizing advance
directives and proxy appointments to maximize
the medical autonomy of persons with diminished
capacity?

• First, and not surprisingly, many patients do
not want to dwell in detail on the prospect of
their future incompetence and death and,
therefore, do not wish to discuss advance direc-
tives (8).

• Second, even when a patient completes a liv-
ing-will type of advance directive, the forms are
often so vague and so open to interpretation
that they are of little guidance when specific
treatment decisions are required.

• Third, a preference expressed during good
health and relatively high levels of function
may not persist; that is, people can change
their minds, and in the real world of clinical
medicine, often do, depending on changing cir-
cumstances (8, 9). This problem points to the
obvious advantage of a proxy being appointed
to work with the doctor in "real time" as cir-
cumstances unfold.

• Finally, in some cases of chronic dementia, pa-
tients may seem to have a pleasant and enjoy-
able quality of life that would ordinarily de-
serve continued medical treatment, but their

548

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

advance directive refuses life-sustaining treat-
ment under such circumstances (9). Should the
doctor treat this patient, or should she instead
have primary loyalty to the document executed
by a younger version of this person under quite
different circumstances?

Similar problems plague the proxy appointment
process: proxies may be unable to distinguish
their own best interests from the patient's wishes
or best interests; many studies have demon-
strated fairly poor concordance of proxy choices
with the patient's stated preferences (10, 11); and
proxies may act in direct opposition to the pa-
tient's treatment directive.

Changes Needed
To Promote Autonomy

To ensure that these well-intentioned efforts
to promote patient self-determination actually re-
sult in decisions that most people would judge to
be in their best interest, several changes are
needed:

1. Advance treatment directives should
stimulate discussion between doctor and patient,
rather than replace it. Obviously, such discus-
sions should occur when the patient is in a stable
condition and not when a patient is being admit-
ted to a hospital for an acute illness.

2. Intensive public and professional educa-
tion on the advantages and limitations of modern
medical technology is required to modify both pa-
tients' and doctors' unrealistic expectations and
hopes.

3. Care must be taken to avoid coercing pa-
tients to complete treatment directives in order to
conserve scarce health care dollars — we must not
ask patients and families and their personal phy-
sicians to ration care in the guise of promoting

their control over medical decisions at the end of
life — that responsibility should rest with demo-
cratically legislated limits on spending and equi-
tably applied medical prognostic criteria for the
use of life-sustaining technologies.

4. Despite our best efforts, the principle of
autonomy is often nearly irrelevant to medical
decision making for frail older persons at the end
of life. As clinicians, we must develop flexibility
to rely on autonomous choices when that is a
meaningful concept for the patient at hand, and
to return to more traditional best-interest criteria
when it is not.

References

1. Cassel CK, Brody J A. Demography, epidemiology and ag-

ing. In: Cassel CK, Riesenberg DE, Sorenson LB, Walsh
JR, eds. Geriatric medicine, 2nd ed. New York:
Springer Verlag, 1990, 17.

2. Moody HR. Ethics in an aging society: old answers, new

questions. In: Ethics in an aging society. Baltimore:
Johns Hopkins University Press, 1992, 1.

3. Schloendorff v. Society of New York Hospitals, 211 N.Y.

125, 127, 129; 105 N.E. 92, 93 (1914).

4. Cruzan v. Director, Mo. Dept. of Health, 110 S. Ct. 2841

(1990).

5. Society for the Right to Die. Refusal of treatment legisla-

tion: a state-by-state compilation of enacted and model
statutes. New York: Society for the Right to Die, 1991.

6. Omnibus Budget Reconciliation Act of 1990. Pub. L. No.

101-508 4206, 4751 (codified in scattered sections of 42
U.S.C., especially 1395), cc, 139 6a (West Supp. 1991).

7. New York Pub. Health Law 2984 (McKinney Supp. 1991).

8. Meier DE. The physician's experience: elderly patients.

Mt Sinai J Med 1991; 58:385-387.

9. Wolf SM, Boyle P, Callahan D, Fins JJ, et al. Sources of

concern about the Patient Self-Determination Act. N
Engl J Med 1991; 325:1666-1671.

10. Seckler AB, Meier DE, Mulvihill M, Paris BEC. Substi-

tuted judgment: how accurate are proxy predictions?
Ann Intern Med 1991; 115:92-98.

11. Uhlmann RF, Pearlman RA, Cain KC. Physicians' and

spouses' predictions of elderly patients' resuscitation
preferences. J Gerontol 1988; 43:M115-M121.

Education, Clinical Services, and Research
On Treating Older People

Long-Term Care

A Teaching Nursing Home:

Ten Years of Partnership between the Jewish Home and Hospital for
Aged and the Mount Sinai School of Medicine

Leslie S. Libow, M.D.

Quality in long-term care depends on a linkage
to education. Thus, there is no true quality nurs-
ing home without an academic component and
there is no true geriatric medicine without a
nursing home program at its center (1).

America's first teaching nursing homes em-
phasized training of medical students, residents,
and fellows, combined with program innovation
and research. These academic nursing homes, as
they were referred to in the start-up years of 1967
to 1977, provided the base for the development of
U.S. -trained geriatricians and for the develop-
ment of the field of geriatric medicine (1-5).
Great impetus was given to this movement by
the far-reaching paper "The Teaching Nursing
Home" (6).

The more recent experience of the Jewish
Home and Hospital for Aged (JHHA) of New York
and the Mount Sinai School of Medicine in devel-
oping a teaching nursing home partnership began
in 1982 (7). The program has encompassed more
than 1,000 senior medical students in an obliga-
tory clerkship and 50 fellows who have completed
their training at the nursing home. Our programs
have led to new clinical approaches and teaching
curricula and to research into a variety of issues
important to long-term care. Interest has been
strong among participating medical students,
graduate physicians, staff, and patients and their

From The Jewish Home and Hospital for Aged and The Mount
Sinai School of Medicine, New York City. Address reprint re-
quests to Leslie S. Libow, M.D., Chief of Medical Services, The
Jewish Home and Hospital for Aged, 120 West 106th Street,
New York, NY 10025.

Presented in part at the 10th year Anniversary Celebra-
tion of the Department of Geriatrics and Adult Development
Partnership with the Jewish Home and Hospital for Aged of
New York, May 1993, New York City.

families, as well as among the Board of Trustees
and administration (5, 8, 9).

Many teaching nursing homes have docu-
mented their experience (9-15) though few, if
any, have described a clerkship with an obliga-
tory rotation of all medical students.

The goals have been to train every medical
student in the art and science of geriatric medi-
cine while improving care for the frail elderly. We
at the nursing home have been mentoring the fel-
lows in their growth as clinicians, teachers, and
researchers, while at the same time developing
strong communication with the administrative
leadership and the Board of Trustees of The Jew-
ish Home and Hospital for Aged and The Mount
Sinai Medical Center.

Education

Geriatrics as a Clinical Specialty. Geriatrics
is partly defined by the phases and places where
its approaches are needed and expressed. Thus, it
is chronic illnesses, functional disabilities, the
prevention of secondary and tertiary complica-
tions that are the focus of geriatrics. Similarly, it
is at the nursing home and the ambulatory clinic
and in home care that the greatest need exists for
input from geriatric medicine.

Of course, the hospital too needs innovation
with regard to its care of the acutely ill frail older
patients. There are some who believe that geriat-
rics is not a clinical-practice specialty, but more
akin to clinical pharmacology or infectious dis-
ease. They see geriatrics as an institutional-based
addition to the current practice of medicine. That
view is too restrictive and fails to encompass the
major clinical responsibilities that define the
shape and substance of the field of geriatrics.

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

551

552

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Components of the Geriatrics Curriculum in
Nursing Home Medicine. Because of the key role
of long-term care in geriatrics, and because of the
willingness of the nursing home and its home-
care program to participate as training sites, we
at The Jewish Home and Hospital and the Mount
Sinai School of Medicine obliged all senior-year
medical students to rotate to the nursing home
and its multitude of institutional and community
programs. The typical components of the two-
week and four-week curriculum, which high-
lights multiplicity of problems, is as follows:

• Functional assessment and intervention: phys-
ical and mental

• Ethical dilemmas, ethics rounds, and other ap-
proaches

• Rehabilitation; diagnosis and treatment

• Home care — visits and teams

• Body of knowledge of long-term care

— Pneumonia: organism; treatments; vaccine
efficacy

— Epidemics: flu; diarrhea

— Patterns of medication use

— Difficult diagnosis of the common problem

• Mentors for research projects

• Daily rounds in a variety of disciplines

• Screening and prevention

Although all students were in their senior year in
medical school, most students confronted several
first-time experiences in this rotation, including
(a) the first house call and home-care team expe-
rience, revealing the power of this nonhospital
approach, and the key role of clinicians other
than physicians (16); (b) the first opportunity to
learn the skills of rehabilitation medicine and of
diagnosing and prescribing for physical disabili-
ties such as hemiplegia, fractured hip, and un-
steady gait (17); (c) the first obligatory clerkship
in which the major educational experience was
not at the acute-care hospital but rather at long-
term care settings (8, 18, 19).

Attitudes of Rotation Graduates. In 1983,
before the arrival of the first senior class obli-
gated to undergo the nursing home rotation, a
delegation of students came to see program orga-
nizers and protested the new curriculum. It was
bad enough, they said, to have the nation's first,
or one of the first, obligatory four-week geriatric
clerkships, but to spend two to four of those weeks
on the nursing home programs was unacceptable.
We clarified their duties, pointed out the unique
experiences available, and did not change the cur-
riculum (7, 18, 19).

More than 10 years and 1,000 students later,
those predictable complaints have receded. Stu-

dent evaluations place this now established geri-
atrics curriculum alongside the classic clerkships
in medicine, surgery, and pediatrics.

JHHA researchers located and mailed ques-
tionnaires to 600 of the 1,000 graduates and re-
ceived 306 anonymously completed question-
naires. Approximately 70% of these clinicians
and former JHHA students favor keeping the
four-week geriatrics curriculum or increasing it
to more time whereas 30% favor decreasing the
clerkship time (Olson E, Libow LS, Chichin E,
"Long-term effects of a mandatory fourth-year
medical student rotation in geriatric medicine,"
unpublished research):

• Would recommend increasing length

of geriatric clerkship 20.9%

• Think length of clerkship should

remain the same 48.6%

• Would recommend decreasing length

of geriatric clerkship 30.5%

This strong support derives from the value of
training in treating older patients as seen by cli-
nicians once they are in residency, fellowship, or
practice.

Research

The teaching nursing home academic staff,
with central input from fellows, chose six re-
search areas — education; rehabilitation, falls,
and restraints; urinary incontinence; infection
and immunology; ethical dilemmas; and Alzhei-
mer's disease and other dementias — as focal in
improving the quality of life of patients in long-
term care. The interests of particular fellows has
led the mentor-fellow partnerships into new areas
of research beyond these six.

Among the many studies completed are
these:

Education. Experiences of students have
been measured and show a statistically signifi-
cant increase in knowledge in comparisons of be-
fore-clerkship and after-clerkship knowledge lev-
els (18, 19). The positive view of graduates has
been described above.

Rehabilitation. Published reports have de-
scribed a falls consultation service (20) and a re-
habilitation unit (17) in the nursing home. More
recently, data have been gathered delineating a
decrease in use of restraints from 51% to 3% at
JHHA, and a similar trend in a national restraint
reduction effort which we have created in collab-
oration with The Commonwealth Fund and Rens-
selaer Polytechnic Institute (21).

Vol. 60 No. 6

A TEACHING NURSING HOME— LIBOW

553

Urinary Incontinence. Fellows are trained in
urodynamics and the assessment of incontinence.
Our studies have shown the remarkably high
prevalence of urinary incontinence and its sur-
prising etiology (detrusor hyper-re flexia) in most
elderly diabetics (22-24).

Infection and Immunology. Jewish Home
and Hospital for Aged studies have shown the ef-
ficacy of the influenza vaccine in decreasing re-
spiratory infection and mortality in the nursing
home population (25). In addition, the high rate of
occurrence of anergy in the frail elderly has been
clarified, and its mechanism is under study.

Ethical Dilemmas. A variety of clinical eth-
ical issues have been studied. One example is the
discordance between patients' wishes and their
caregivers' beliefs about their wishes (26). In ad-
dition, an ethics rounds (27) and a center on eth-
ics in long-term care (28) have been created.

Alzheimer's Disease and Other Dementias.
Alzheimer's disease is being studied in collabora-
tion with investigators and laboratories at The
Mount Sinai Medical Center. JHHA is unusual as
a long-term care center in having on-site autop-
sies. The Alzheimer's disease research focus is on
correlating clinical cognitive changes with ana-
tomic and biochemical changes. We are also
studying the relationship of Alzheimer's disease
to changes in the normal aging brain.

Obstacles and Challenges

In 1982 the curriculum committee of Mount
Sinai School of Medicine was reluctant to cooper-
ate with the Dean's commitment to the four-week
clerkship in geriatric medicine. Thus, resistance
existed even in the face of the dedicated plan to
create this innovation. This expected response
was surmounted. It was our decision to heavily
utilize long-term care as a basis for education.
Over the ensuing years the hospital component of
the geriatric curriculum developed strongly too.

The nursing home staff had the anticipated
discomfort and distrust of the medical school and
hospital faculty. Our approach was to assign med-
ical school faculty members to primary teaching,
clinical, and administrative responsibilities at
the nursing home, so that they would not be out-
siders. This has produced a collegial working re-
lationship, the needs of the nursing home always
remaining primary while faculty identification
with the nursing home patients and programs is
strong.

Overt discomfort at the nursing home by
some faculty in the geriatrics department was a
bit surprising. We ultimately identified as JHHA

faculty those individuals comfortable with and
committed to long-term care.

Geriatric medical programs at the nursing
home attract philanthropic and government sup-
port. The combined efforts of the JHHA and the
Mount Sinai School of Medicine established an
endowed professorship in long-term care (the
Greenwall Professorship). JHHA established an
Ethics Center in Long Term Care (Schimper,
Greenberg & Zicklin Foundation); a national Re-
straint Reduction Project (Commonwealth Fund),
and a rehabilitation program (the Brookdale
Foundation and Federation of Jewish Philanthro-
pies). In addition, since 1987 government funds
have paid for a sizable portion of the fellowship
salaries.

Summary and Conclusions

The JHHA is moving forward into new areas
of pursuit within long-term care, including stud-
ies of the pneumonia vaccine; the cause of Alzhei-
mer's disease; the difference between the brain of
the Alzheimer's patient and the normal aging
brain; new patterns and approaches in pharmacy
and pharmacology; and the new nursing home of
the 21st century in a changing health care envi-
ronment.

Students and graduate physicians will con-
tinue to strengthen their clinical identity by vir-
tue of this exposure, and those of all ages will
benefit from the research and advances in patient
care coming out of this program.

References

1. Libow LS. Geriatric medicine and the nursing home: a

mechanism for mutual excellence. Gerontologist 1982;
23:134-144.

2. Libow LS. A fellowship in geriatric medicine. J Am

Geriatr Soc 1972; 20:580-584.

3. Libow LS. A geriatric medical residency program: a four-

year experience. Ann Intern Med 1976; 85:641-647.

4. Williams TF, Izzo AJ, Steel RN. Innovations in teaching

about chronic illness and aging in a chronic disease
hospital. In: Clark DW, Williams TF, eds. Teaching of
chronic illness and aging. Publication (NHH), 75-786,
Dept. of Health, Education & Welfare, 1975, pp. 21-30
and Appendix A.

5. Libow LS. From Nascher to now: seventy-five years of

United States geriatrics. J Am Geriatr Soc 1990; 38:79-
83.

6. Butler RN. The teaching nursing home. JAMA 1981;

245(14):1435-1437.

7. Butler RN. Mount Sinai School of Medicine. In: Schneider

EL, Wedland CJ, Zimmer AW, et al., eds. The teaching
nursing home. NY: Raven Press, 1985, 99-104.

8. Libow LS. The teaching nursing home: past, present, and

future. J Am Geriatr Soc 1984; 32:598-603.

9. Duthie EH, Prieter B, Gambert SR. The teaching nursing

home— one approach. JAMA 1982; 1787-1788.

554

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

10. Schneider EL. Teaching nursing home. N Engl J Med

1983; 308:330-337.

11. Aiken LH, Mercy MD, Lynaugh JE, et al. Teaching nurs-

ing homes: prospects for improving long-term care. J
Am Geriatr Soc 1985; 3:196-201.

12. Weiland D, Rubenstein LT, Ouslander JG, et al. Organiz-

ing an academic nursing home: impacts on institution-
alized elderly. JAMA 1986; 25:2602-2627.

13. Jahnigen DW, Kremer AM, Robbins LJ, et al. Academic

affiliation with a nursing home: impact on patient out-
come. J Am Geriatr Soc 1985; 33:472-^78.

14. Rubenstein LT, Wieland D, Pearlman RA, et al. Growth of

the teaching nursing home: the Veterans Administra-
tion experience. J Am Geriatr Soc 1990; 38:73-78.

15. Powers JS, Burger C, Manian FA, et al. A teaching nurs-

ing home: the Vanderbilt experience. South Med J
1986; 79:267-271.

16. Tideiksaar R, Libow LS, Chalmers M. House calls to older

patients: the medical student experience. Pride 1985
(Summer); 4:3-8.

17. Adelman RD, Marron K, Libow LS, Neufeld RR. A com-

munity-oriented geriatric rehabilitation unit in a nurs-
ing home. Gerontologist 1987; 27(2):143-146.

18. Murden RA, Meier DE, Bloom PA, Tideiksaar R. Re-

sponses of fourth-year medical students to a required
clerkship in geriatrics. J Med Educ 1986; 61:569-578.

19. Fields SD, Jutagir R, Adelman RD, et al. Geriatric edu-

cation: Part I. Efficacy of a mandatory clinical rotation

for fourth-year medical students. J Am Geriatr Soc
1992; 40(9):964-969.

20. Neufeld RR, Tideiksaar R, Yew E, et al. A multidisci-

plinary falls consultation service in a nursing home.
Gerontologist 1991; 31(1):120-123.

21. Dunbar JM, Cohen CE, Neufeld RR, et al. Removing re-

straints: training, consulting and educating. (Gerontol-
ogist 1992 (Special Issue 2); 32:48-49.

22. Starer P, Libow LS. Cystometric evaluation of bladder

dysfunction in elderly diabetic patients. Arch Intern
Med 1990; 150:810-813.

23. Starer P, Libow LS, Hsu MA. The association of inconti-

nence and mortality in elderly nursing home patients. J
Med Direct 1992; 2:49-51.

24. Starer P. Cystometric evaluation of elderly nursing home

patients with indwelling urinary catheters. Arch Ger-
ontol Geriatr 1992; 15:79-86.

25. Gross PA, Quinnon GV, Rodstein M, et al. Association of

influenza immunization with reduction in mortality in
an elderly population. Arch Intern Med 1978; 148:562-
565.

26. Michelson C, Mulvihill MN, Hsu MA, Olson E. Eliciting

medical care preferences from nursing home residents.
Gerontologist 1991; 31(3):358-363.

27. Libow LS, Olson E, Neufeld RR, et al. Ethics rounds at the

nursing home: an alternative to an ethics committee. J
Am Geriatr Soc 1992; 40:95-97.

28. Olson E, Chichin ER, Libow LS, et al. A center on ethics

in long-term care. Gerontologist 1993; 33:269-274.

Ethical Issues in the Nursing Home

Ellen Olson, M.D.

Abstract

The nursing home is the site of many ethical dilemmas. The majority of these are linked
to the high prevalence of dementia and other disorders that affect decisional capacity, thus
dictating the need for surrogate decision-making. Even when nursing home residents have
utilized advance directives to guide their future care, many issues remain in the decision-
making process. This article explores the challenges of medical decision-making for a frail
elderly population.

A PRIMARY FOCUS of ethlcal discussion in the nurs-
ing home is the nursing home resident without
capacity. Historically, the family and the primary
care physician were the decision-makers for this
group of individuals. Although some states con-
tinue to recognize this approach and have adopted
legislation to support it, the majority of states —
including New York — do not currently do so.

The Legal Climate. Major court decisions
over the past several years have, in fact, sup-
ported a trend away from family decision-making
to a decision-making standard that is resident-
focused and stresses the importance of advance
directives (living wills and proxy designations)
and "substituted judgement" decisions by surro-
gate decision-makers. In other words, surrogate
decisions should, to the extent possible, mirror
the decisions the patients themselves would make
if they were capable of doing so.

This trend culminated in 1990 in the Cruzan
decision, in which the Supreme Court of the
United States reaffirmed patient autonomy and
the right to refuse treatment. However, at the
same time, it rendered families powerless to
speak for their incapacitated family members in
states that required clear-cut directives from the

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, and The Jewish Home and
Hospital for Aged. Address reprint requests to Ellen Olson,
M.D., The Jewish Home and Hospital for Aged, 120 W. 106th
Street, New York, NY 10025.

patients prior to their incapacity (1). The Appel-
late Court of New York State had reached a sim-
ilar decision in 1988 in the O'Connor case, in
which the precedent of a standard of "clear and
convincing" evidence was adopted before life-sus-
taining treatments could be removed from pa-
tients incapacitated in decision-making (2). Un-
fortunately, this standard is difficult to meet
without written directives.

In response to the Cruzan decision, a proxy
law went into effect in New York State in 1991,
allowing people with capacity to appoint a deci-
sion-maker in the event of incapacity. The pas-
sage by Congress in 1990 of the Patient Self-De-
termination Act encouraged states to adopt
legislation that allows the use of advance direc-
tives for health care. However, this still left open
the question of decision-making for residents who
lacked capacity and had never made use of an
advance directive. Equally important, it did not
address who should make treatment decisions for
this vulnerable group of individuals.

In an attempt to deal with this issue, the New
York State Task Force on Life and the Law has
proposed legislation to address surrogate deci-
sion-making for patients without capacity. How-
ever, the proposal is in the early stages of the
legislative process, and how long it will be before
the legislation becomes reality is not clear. Mean-
while, making life-sustaining treatment deci-
sions for residents without capacity in New York
State and in other states in which such laws do

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

555

556

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

not exist is extremely difficult. In these states,
such decisions can only be made when adequate
advance directives are in place prior to the loss of
capacity or where other forms of "clear and con-
vincing" evidence exist.

This reliance on advance directives means
that their strengths and weaknesses affect many
of the ethical issues faced in the nursing home
setting, as well as other situations in which cli-
nicians turn to these documents. Even when re-
quests to limit treatment for people without ca-
pacity are supported by valid directives, much
work remains before any conclusion can be
reached and a plan of treatment or nontreatment
implemented. This situation also holds, however,
when requests are made to limit treatment by a
person with decision-making capacity.

Treatment Decision-Making:
The Process

Medical Evaluation. How does one approach
difficult treatment decisions in the nursing
home? At the heart of every discussion to limit
treatment is the need for a sound and thorough
medical evaluation. Some older people may wish
to forgo life-sustaining treatments just because
they are experiencing pain or discomfort or pro-
found functional losses resulting from acute or
chronic illnesses. The often confusing presenta-
tion of disease in the elderly in general, combined
with the tendency to overlook illness in the elder-
ly because it is attributed to "old age," may lead
to some people being considered "terminally or
irreversibly ill" when they really have a revers-
ible medical problem whose cure could change
their whole outlook on continued life.

Advance directives invoked for incompetent
residents are often linked to stated conditions,
such as "terminally ill" or "irreversible mental
impairment." The determination must be made
objectively to insure that the person indeed meets
the conditions of the directive. Clinicians are of-
ten too quick to draw conclusions in either direc-
tion. On the one hand, they may deem the situa-
tion hopeless when it is not. On the other hand,
they may disagree with residents or families over
the futility of the situation. Again, achieving this
comprehensive and objective review of a resi-
dent's actual medical status is really the first
challenge in these cases. And, unfortunately, this
can often be complicated by clinicians' ambiva-
lence toward refusal of treatment.

Determining Capacity. The next major chal-
lenge faced in the nursing home is the determi-

nation of decision-making capacity or compe-
tence. A resident must be deemed competent to
refuse treatment, and in the case of proxy deci-
sions, a resident must be deemed incompetent or
incapable of participating in the decision process
before the assigned agent can speak for him or
her.

One cannot talk about capacity or compe-
tence without first talking about informed con-
sent. The criteria for informed consent are: (a) it
must be voluntary; (b) it must be given by a per-
son who is competent — the person must have the
mental capacity to make and communicate auton-
omous decisions; and (c) it must be informed —
patients must be informed of all their options and
the risks and benefits of each, and they must ap-
pear to understand the information (3).

The questions to be asked in assessing com-
petency or capacity include whether the person
can make and communicate (by spoken word or
otherwise) decisions regarding his or her ovm life,
can offer any rational reasons for the decision,
and understands the likely risks and benefits of
the options presented and that those risks and
benefits apply to him or her (4).

In the nursing home, we most often run into
situations of partial competence. The rule of
thumb is that the degree of competence or capac-
ity needed is directly related to the gravity of the
decision. When a sick person is refusing life-sus-
taining therapies, the level of competence or ca-
pacity should be quite high, but need not be what
is considered normal. Studies have shown that
even when older people do not do as well as
younger people in understanding and recalling
information (for example, the information de-
scribing a procedure for which consent is needed),
they still make reasonable decisions that reflect
their ability to use the information they do pro-
cess to assess benefits and risks (5).

Influence of Depression. Another factor that
may complicate the determination of capacity in
the nursing home is the presence of depression.
As much as 24% of nursing home residents may
suffer from depression (6). Although depression
has not been found to affect overall comprehen-
sion and the ability to give informed consent (7),
many believe it influences, at least to some de-
gree, a person's decision on the use of life-sustain-
ing treatment.

Many elderly people in nursing homes suffer
from severely disabling conditions that entail loss
of function and independence and little to no hope
of recovery, and these patients may have a con-
comitant depression. It has been argued that ag-

Vol. 60 No. 6

ETHICS AND DIMINISHED CAPACITY— OLSON

557

gressive treatment should be instituted for de-
pression in elderly patients, especially those who
are medically ill, because depressive disorders are
often associated with an increased mortality rate
in such patients. However, this should not be con-
sidered the only acceptable option (8).

Lee has also argued that although depression
may alter one's capacity to make treatment deci-
sions, it does not preclude the ability to make
some decisions. She expresses a concern that cli-
nicians are often reluctant to respect requests to
limit treatment from depressed individuals be-
cause the physician perceives limitation of treat-
ment to be assisting a suicide. People with poten-
tial for recovery from a period of significant
physical or emotional pain or disability should
clearly be given the time and treatment they need
to facilitate such a recovery. Many older persons'
situations will not improve, however, and as Lee
states, "When suffering is unlikely to abate, a de-
cision that death is preferable to life may not nec-
essarily be unreasonable" (8). This is clearly a
topic that warrants more discussion and research.
However, no one has reflected more on a person's
life than the person who has lived it, and others
must not be too quick to take away the ability to
control that life for dubious reasons.

The Role of Advance Directives

Once the medical evaluation is complete and
the decision-making capabilities are determined,
the next goal is to ascertain the wishes of the
patient. This is relatively straightforward for a
person with capacity; however, approximately
50% of nursing home residents have some form
of mental impairment that may preclude the abil-
ity to provide this information (9). Here is where
advance directives and prior discussions with
friends and family become paramount.

The two most commonly recognized forms
of advance directives are the living will and the
durable power of attorney for health care. The
durable power of attorney is equivalent to the
proxy designation in New York State. The main
problem with advance directives in general is
that only a minority of people use them (10).
Therefore, most requests by family members to
withhold or withdraw treatment from nursing
home residents are not supported by directives.
This necessitates the exploration of other forms
of evidence, such as recollections of conversa-
tions with the resident when he or she had capac-
ity about end-of-life treatment or recollections
of how he or she responded to any similar medical

conditions and treatment decisions of friends or
relatives.

Problems with Living Wills. Even when liv-
ing wills exist, many of them — especially older
documents — tend to be vague and generally do
not satisfy the standard of "clear and convincing"
evidence. Documents several years old often raise
concerns about whether the person may have
changed his or her mind about refusing treat-
ment, especially when the treatments seem rela-
tively commonplace, such as antibiotics.

In a recent Hastings Center Report, John
Robertson expressed his belief that many physi-
cians are justified in not taking living wills seri-
ously. When we become incompetent, we really
become different people, with different needs and
interests that we really cannot appreciate ahead
of time. Accordingly, directives written by com-
petent persons for periods of incompetence should
be suspect. Robertson suggests that only family
members and physicians can assess what our
needs and interests are as they observe us in our
incapacity (11).

However, this is not the generally accepted
view. Although it is recommended that living
wills be reviewed periodically and that review be
somehow documented, many people spend years
in a state of incapacity and can no longer exercise
this safeguard. The fact remains that for those
who have one, the living will is the best evidence
we have of their wishes. We should not presume
that someone may have changed his or her mind
about care, absent any evidence to that fact.

Advantages of Proxy Appointments. The
limitations of living wills — the applicability of
patient preferences to current situations, vague
wording, and so on — have led many people to be-
lieve that, when filling out an advance directive,
a proxy designation is preferable, for it is more
flexible, empowering someone to address the va-
garies inherent in living wills. Although it is an-
ticipated that more and more people will utilize
these directives, the evidence is ample that the
current generation of elderly persons will con-
tinue to rely on family for decision-making, with-
out understanding that their families are not nec-
essarily empowered to speak for them when they
can no longer speak for themselves (12). Even
when proxy documents are in place, many studies
have challenged whether surrogate decision-
making can ever achieve the goal of substituted
judgment, absent the coexistence of more detailed
documentation of wishes (13, 14). Nevertheless,
trying to ascertain the wishes for treatment of a
nursing home resident remains the ideal.

558

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

Artificial Nutrition
And Hydration

Clearly, the request in the nursing home that
invokes the most debate and turmoil among staff
is to withhold or withdraw nutrition and hydra-
tion administered via a nasogastric or other type
of feeding tube. Some of the debate hinges on le-
gal requirements that nursing homes provide ad-
equate nutrition and hydration to all nursing
home residents. In New York State, for example,
failure to do so, except under carefully detailed
circumstances, is regarded as resident abuse. Be-
yond that is the feeling of many health care pro-
viders that in almost all circumstances, being fed
is routine care that should never be refused or
denied. In spite of court decisions, including the
Cruzan decision, which characterized tube feed-
ings as "treatments" that could be refused like
any other treatment, tube feedings are often not
seen as such (1).

In addition to presenting a moral dilemma,
consideration of tube feedings again has to in-
volve a rigorous medical evaluation. In an elderly
person, loss of appetite is one of the most com-
mon symptoms of a myriad of disorders, ranging
from fecal impaction and urinary tract infection
to drug reactions. Again, staff members some-
times too easily feel there are no alternatives to
tube feeding, and families sometimes too easily
feel that the resident warrants no further treat-
ments, even when the cure may be relatively sim-
ple, and adequate evaluation and treatment
will restore the resident to begin eating on his or
her own again. A prompt and thorough medical
evaluation is necessary in the first case before
tube feedings are instituted, whereas in the sec-
ond situation a time-limited trial can often be
helpful.

In situations where the "clear and convinc-
ing" standard is met and tube feedings may legit-
imately be refused, uncertainty may remain
about the resident's status and the family or the
agent's level of comfort with the request to with-
hold tube feedings. The ethically acceptable ap-
proach is to try treatments that may prove bene-
ficial and withdraw them if they fail to achieve
their goals. There is no ethical or moral distinc-
tion between withholding and withdrawing a
treatment, nor is there a legal distinction. Ethi-
cists feel the general rule is. When in doubt, treat,
then deliberate and withdraw the treatment if it
is deemed either ineffective or inappropriate. In
most cases, the psychological burden this places
on staff and family can sometimes be outweighed
by the knowledge that the treatment was tried
and did not achieve its goal.

Conclusion

Even when a nursing home resident's wishes
are clear and a medical evaluation reveals the
resident to be in a condition in which wishes to
forgo treatment should be respected, several
other issues must be addressed. Caregivers, in
particular physicians and nursing staff, may have
personal disagreements with the choices made by
the resident. It is imperative that all staff who
will be involved in the implementation of a deci-
sion to withhold or withdraw treatment from a
nursing home resident, or from any patient, are
included in the decision-making process. Even if
they disagree with the final outcome, knowing
how a decision was reached and knowing that it
truly reflects the wishes of the resident can only
make providing the comfort-care needed easier.

More important, whenever possible, staff
who will be faced with difficult decisions about
patients at the end of life should be educated
about ethical issues and involved in the develop-
ment of institutional policies that guide decisions
to withhold or withdraw treatments.

Finally, staff also need to be educated in com-
fort-care issues and also need emotional support
to assist them as they care for residents for whom
treatments are being withheld. Most nursing
home staff develop close relationships with the
residents for whom they care, and this, coupled
with perceived professional obligations to always
support life and with conflicts with decisions
made by others to forgo treatments, can make
caring for these residents very difficult. Empha-
sizing that respect for resident autonomy is one of
the greatest gifts staff can give to those for whom
they care may ease their concerns.

References

1. Cruzan V. Director, Missouri Department of Health, 110

S. Ct. 2841, 2855-56 (1990).

2. In re O'Connor, 1988; 72 N.Y.2d 517, 531 N.E.2d 607, 534

N.Y.S.2d 886.

3. Kapp MB. Informed consent: an ongoing process of com-

munication. Brown Univ Long-Term Care Lett 1991;
(July 11);8-10.

4. Kapp MB. Decision making by and for nursing home res-

idents: a legal view. Clin Geriatr Med 1988; 4(3):667-
678.

5. Stanley B, Guido J, Stanley M, Shortell D. The elderly

patient and informed consent. JAMA 1984; 252:1302-
1306.

6. Heston LL, et al. Inadequate treatment of depressed nurs-

ing home elderly. J Am Geriatr Soc 1992; 40(11):1117-
1122.

7. Stanley B, Stanley M, Guido J, Garvin L. The functional

competency of elderly at risk. Gerontologist 1988; 28:
53-58.

8. Lee M. Depression and refusal of life support in older peo-

ple: an ethical dilemma. J Am Geriatr Soc 1990; 38:
710-714.

Vol. 60 No. 6

ETHICS AND DIMINISHED CAPACITY— OLSON

559

9. Harper MS, Lebowitz B. Mental illness in nursing homes:
agenda for research. Rockville, Md: National Institute
of Mental Health, 1984.

10. Greco P, Schulman K, Lavizzo-Mourey R, Hansen-

Plaschen J. The Patient Self-Determination Act and
the future of advance directives. Ann Intern Med 1991;
115(8):639-643.

11. Robertson J. Second thoughts on living wills. Hastings

Cent Rep 1991; 21(6):6-12.

12. High D. All in the family: extended autonomy and expec-

tations in surrogate health care decision-making. Ger-
ontologist 1988; 28:46-51.

13. Uhlman RF, Pearlman RA, Cain KC. Understanding of

elderly patients' resuscitation preferences by physi-
cians and nurses. West J Med 1989; 150:705-707.

14. Seckler AB, Meier DE, Mulvihill MN, Paris BEC. Substi-

tuted judgement: how accurate are proxy predictions?
Ann Int Med 1991; 115(2j:92-98.

Nursing Staff Attitudes on the Use of
Physical Restraints in a Teaching

Nursing Home

Janet Michello, PhD, Richard R. Neufeld, MD, Michael Mulvihill, Dr. PH, and Leslie S. Libow, MD

Staff in hospitals and nursing homes in the
United States, unlike staff in European facilities,
routinely use physical restraints on patients (1).
Before the Omnibus Budget Reconciliation Act
(OBRA) became effective in October 1990,
throughout the United States 500,000 older
Americans were tied to their hospital or nursing
home beds and chairs (2). About half of all New
York State nursing home residents were physi-
cally restrained at least part of each day (1). The
OBRA-defined federal regulations on nursing
homes on the use of restraints in health-care fa-
cilities specified the rights of each resident to be
free from physical or chemical restraints imposed
for purposes of convenience and not required for
treating a resident's symptoms (3).

Prolonged immobilization of the elderly con-
tributes to such physical disabilities as disordered
motor function, contractures of joints, edema, de-
cubitus ulcers, and incontinence; it also contrib-
utes to many behavioral outcomes, such as de-
pression, social withdrawal, apathy, and anxiety
(4). Strumpf and Evans interviewed 20 restrained
hospitalized elderly patients and found their re-
sponses to be primarily fear, demoralization, dis-
comfort, anger, humiliation, and denial (5).
McHutchion and Morse note that initially re-
strained patients are angry and sometimes hos-
tile and eventually develop progressive physical
and emotional deterioration (6).

From the Departments of Sociology at the University of Ak-
ron-Wayne College, Orville, OH (JM), and The Jewish Home
and Hospital for Aged, New York, NY (RRN, MM, LSD. Ad-
dress reprint requests to Janet Michello, Ph.D., at the Uni-
versity of Akron-Wayne College, Department of Sociology,
10470 Smucker Road, Orville, OH 44667. Copies of the ques-
tionnaires can also be obtained there.

560

Despite awareness of the negative outcomes
of physical restraints, they continue to be used in
the United States. Some investigators have sug-
gested that few alternatives exist (5) and that re-
straints are most commonly used to protect pa-
tients (2). Others have classified the reasons for
restraining the elderly as "external" and "inter-
nal." According to McHutchion and Morse, exter-
nal situations include those in which staff feel
they have little control, experience pressure from
families, and fear being liable should a fall occur;
staffs internal or intrinsic reasons for using re-
straints include beliefs about lack of alternatives
and safety issues and perceptions that using re-
straints saves time (6). Similarly, Williams con-
siders such factors as staff attitudes, staffing ra-
tios, and ideas about risk being correlated with
restraint use (1).

This study assesses the attitudes of a large
population of nursing assistants and licensed
nursing staff toward using physical restraints in
a long-term-care facility and explores staff per-
ceptions of the effects of restraints. The study hy-
pothesizes that intrinsic and extrinsic factors are
associated with recommendations to apply physi-
cal restraints in various clinical situations. In ad-
dition, the potential effect of formal nursing edu-
cation is explored.

Setting, Study Subjects,
And Methods

Setting. The Jewish Home and Hospital for
Aged (JHHA) is the teaching nursing home affil-
iated with The Mount Sinai Medical Center in
New York City. There are two campuses: a Man-
hattan division consisting of 514 beds, and a
Bronx division of 816 beds. The academic faculty

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

NURSES' ATTITUDES TO RESTRAINTS— MICHELLO ET AL.

561

at the JHHA hold academic appointments in the
Department of Geriatrics and Adult Development
at Mount Sinai Medical School. Geriatrics fellows
and fourth-year medical students do clinical rota-
tions at JHHA.

As of May 1990 the nursing home residents
had an average age of 83 years (Table 1); about
one-third were physically restrained.

Methods. In this study, physical restraints
are defined as any device that binds or surrounds
the resident and is used to restrict movement. Ex-
amples include mittens, waist restraints, criss-
cross belts and straps to tie down hands, and
chairs that prevent rising. For this study, bed
rails were considered possible accident hazards
but not physical restraints.

Nursing staff include all personnel involved
in the direct care of the nursing home residents:
licensed staff and nursing assistants (aides and
orderlies).

A questionnaire was developed consisting of
40 items on extrinsic and intrinsic factors that
potentially contribute to restraint use; the per-
ceived effects of restraint use; and situations in
which physical restraint use might be recom-
mended. Nursing staff were asked to circle the
appropriate number on a 5-point Likert scale (1,
strongly agree; 2, agree; 3, uncertain; 4, disagree;
5, strongly disagree).

A pilot study was conducted on two units that
had minimized restraint use within the year prior
to the study and on two units that were in the
process of minimizing restraint use. The ques-
tionnaire was pretested in the pilot study and re-
fined as a result.

Extrinsic factors that potentially contribute
to restraint use were operationalized as concerns
about the job, the environment, and safety. In-
trinsic factors that potentially contribute to re-
straint use were operationalized as attitude to-
ward restraint use and toward residents. The
perceived effects of restraint use were operation-
alized as physical, emotional, and behavioral ef-
fects and relationship to use of psychotropic med-
ications.

In addition, a series of seven clinical situa-
tions was presented; respondents were asked to
indicate, on a 5-point Likert scale, the likelihood
that they would recommend the use of physical
restraints under various conditions (for example:
"I would recommend the use of physical restraints
for the following: a resident who is confused; a
resident who physically hurts himself or others; a
resident who pulls out an intravenous or nasogas-
tric tube . . ."). Each item was examined sepa-
rately in comparing attitudes of licensed nurses

TABLE 1

Characteristics of Residents of The Jewish Home and

Hospital for Aged, May 1990

Manhattan

Bronx

wlldl aL/LCl lOLl^

Total no.

514

816

No. of SNF

476

646

Average age

83 years

Average length of stay

2.3 years

2.9 years

% male

20%

18%

% temale

80%

82%

% physically

restrained

32%

35%

% taking psychotropic

medication

40%

47%

Most common diagnoses

OMS/dementia

on admission

Post-status

Coronary

stroke

artery

disease

TABLE 2

Characteristics of Nursing Staff of The Jewish Home and
Hospital for Aged

Characteristic

Percent)

Job Title

Licensed staff

31%

Nursing assistants

186

69%

Shift

Day

142

52%

Evening

29%

Night

19%

Education

Some high school

10%

High school grad

37%

Some college

20%

2-year college degree

15%

4-year college degree

Graduate degree

10%

Years employed at facility

<2 years

16%

2—4 years

15%

^5 years

167

69%

Place of birth

United States

31%

Outside U.S.

117

69%

to those of nursing assistants. To obtain a global
view of the tendency to use restraints, responses
to the seven situations were added together. By
using multiple regression, this index of overall
tendency, which took on values ranging from 7 to
35, served as the dependent variable for examin-
ing which characteristics and attitudes were most
strongly related to the tendency to use restraints.

The 40-item questionnaire was distributed at
various staff meetings to all nursing personnel on
all three shifts (day, evening, and night) at both
the Manhattan and Bronx facilities. One investi-

562

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

TABLE 3

Likert Scores} of Nursing Home Staff by Job Title on Factors Contributing to Attitudes on Restraint Use (Mean ± SD)

Contributing factors

Licensed (N = 82)

Assistant (N = 186)

All staff (N = 268)

Extrinsic

Job concerns

Increased work load

Risk of liability

More staff

Closer supervision

Feel better about job
Environment

Layout of unit suitable

More alarm/adaptive devices

More activities
Safety concerns

Safety of residents jeopardized

More falls of residents

Staff harmed

Residents would wander

Residents feel safer restrained

Intrinsic

Attitude to restraint use
Invasion of basic right
Even if risky, worth a try
Restraints in use necessary
Residents never restrained

Attitude to residents
Aware of residents' needs
View job as caring/helping
Encourage resident self-help

Perceived effects of restraint use

Physical

Interfere with medical treatment

Debilitating conditions
Emotions of residents

Angry

Fearful

Humiliated

Discomfort

Helpless

Less calm

Trapped
Behavior of residents

Less agitated

More dependent

Calm

Use of psychotropic medications
If taking, likely to be restrained
Given less if restrained

2.3 ± 1.1

2.6 ± 1.3

1.7 ± .89
1.3 ± .48

3.1 ± .95

3.0 ± 1.2
1.6 ± 6.1
1.9 ± .83

2.3 ± 1.1

2.2 ± .98

3.6 ± .99

2.7 ± 1.1
3.5 ± .89

2.9 ± 1.1
2.1 ± .67

2.5 ± 1.0
3.9 ± 1.0

1.6 ± .64

1.6 ± .74

1.7 ± .81

3.8
2.6

.85
1.1

2.3 ± .91
2.6 ± 1.0
2.5 ± .87
2.3 ± .91
2.3 ± .87

2.2 ± .79

3.4
2.4
2.6

.87
1.1
1.0

3.4 ± 1.0
3.6 ± .94

2.5 ± 1.3

2.3 ± 1.4
1.9 ± 1.1
1.5 ± .67*
3.2 ± 1.0

2.8 ± 1.3
2.0 ± l.Ot

2.0 ± .91

2.2 ± 1.1

1.1 ± 1.1

3.4 ± 1.1

2.5 ± 1.1

3.3 ± 1.1

2.7 ± 1.3
2.3 ± 1.3*

2.5 ± 1.2

3.8 ± .94

1.8 ± .83

1.6 ± .75
1.6 ± .70

3.7 ± 1.0

2.8 ± 1.1

2.2 ± 1.1
2.6 ± 1.1

2.4 ± 1.1

2.1 ± 1.0

2.2 ± .94

2.5 ± .95*

3.5 ± 1.1
2.8 ± 1.2t
2.5 ± 1.1

2.5 ± l.lt

3.3 ± 1.1

2.4 ± 1.2

2.4 ± 1.4

1.8 ± 1.0

1.5 ± .65
3.2 ± .97

2.9 ± 1.2
1.9 ± .92

2.0 ± .87

2.2 ± 1.0

2.1 ± 1.1

3.5 ± 1.0

2.6 ± 1.1

3.3 ± 1.1

2.8 ± 1.2

2.3 ± .90

2.4 ± 1.1

3.9 ± .92

1.7
1.6
1.6

.75
.71
.72

3.7 ± .94

2.8 ± 1.1

2.2 ± 1.0

2.6 ± 1.1

2.4 ± .97

2.2 ± .98

2.3 ± .94
2,6 ± 1.0
2.2 ± .90

3.7
2.7

1.0
1.2

2.8 ± 1.1
3.4 ± 1.1

* <.05; t <.001.

t 1, Strongly agree; 5, strongly disagree.

gator attended various shift meetings to explain
and distribute questionnaires. All nursing per-
sonnel working at that time were included in the
study if they worked on the units where efforts to
minimalize restraints had not yet occurred. Sur-
veys were also distributed at in-service education
meetings. Per diem staff, "floating" staff, and
nursing personnel on vacation at the time of the
survey distribution were not included. Since En-
glish is a second language for many respondents,

the questionnaire items were read to a consider-
able number of them.

The data for this study come from the re-
sponses of 268 nursing personnel (Table 2), 69%
nursing assistants and 31% licensed staff, all
working on units where physical restraints are
used; none had been exposed to restraint-free ef-
forts. All shifts are appropriately represented.
About 90% were at least high school graduates,
and one-third had at least a two-year college de-

Vol. 60 No. 6

NURSES' ATTITUDES TO RESTRAINTS— MICHELLO ET AL.

563

gree. Not all staff completed the questions on ed-
ucation level and place of birth. The majority of
staff (697c) had been employed at the JHHA five
years or more. Most were born outside the United
States, the greatest percentage originating in the
West Indies.

Results

Attitudes. Most nursing staff believe that if
the use of physical restraints were reduced, their
work load would increase, they would be at
greater risk of liability, units would require more
staff, and residents would require closer supervi-
sion (Table 3). They are uncertain about whether
they would feel better about their jobs.

Most agree that more alarm devices and ac-
tivities would be needed if use of physical re-
straints were reduced. Staff are uncertain if the
layout of the units where they work is suitable for
restraint-free residents to move around in. Most
also agree that resident safety would be jeopar-
dized if restraint use were reduced, because resi-
dents would fall more frequently, but staff are
uncertain if residents would be more likely to
wander. They are also uncertain about whether
residents feel safer when restrained. They do not
believe that residents would more easily harm
staff if restraint use were reduced.

Most agree that reducing restraint use is
worth a try, even if these efforts may be risky;
however, they view the majority of restraints cur-
rently in use as necessary, because there are cir-
cumstances in which residents may require phys-
ical restraints. The majority of staff consider
themselves aware of residents' needs and view
caring for and helping residents and encouraging
them to help themselves as the job of nursing
staff.

On the perceived effects of restraint use, most
staff feel that restraints can be reduced without
interfering with medical treatment, but they are
uncertain whether residents develop more debil-
itating conditions when physically restrained.

Most agree that residents become angry and
feel humiliated, uncomfortable, helpless, and
trapped when restrained, but are unsure if resi-
dents become fearful. Most believe that residents
would be less agitated if restraints were reduced
but are uncertain whether residents are more de-
pendent when physically restrained or calmer
when restraint-free.

Survey responses indicate that staff are un-
certain about whether residents taking psycho-
tropic medications would be more likely to be re-
strained. They are also uncertain if residents

would be given more psychotropic medications
should restraint use be reduced.

Overall, the attitudes and perceptions of li-
censed staff are quite similar to those of nursing
assistants (Table 3). However, licensed staff tend
to have fewer concerns about resident safety and
are also less likely to view a reduction in restraint
use as interference with the medical treatment of
residents.

Although the attitudes and perceptions of
nursing assistants are similar to the views of the
licensed staff whatever the situation or status of a
particular resident (Table 4), questionnaire re-
sponses indicate that, given the opportunity,
nursing aides are more inclined to use restraints
in all situations.

Multiple regression analyses identified nurs-
ing assistant status as more strongly associated
with a tendency to recommend restraint use than
licensed status; this status was the factor most
predictive of a tendency to use restraints. Other
such predictors were working the night shift;
fearing being more easily harmed; viewing the
job role as primarily taking care of and helping
residents; believing residents' safety would be
jeopardized by reduction in restraints; and being
more apt to restrain residents taking psychotro-
pic medications.

Discussion

This study supports the hypothesis that in-
trinsic (attitudes toward restraint use) and ex-
trinsic (safety, environmental, and job concerns)
factors contribute to the nursing staffs desire to
use physical restraints. In addition, the staff be-
lieve that restraints contribute to negative phys-
ical, behavioral, and emotional changes in resi-
dents.

TABLE 4

Likert Scores§ of Nursing Staff by Job Title on
Recommendations for Restraint Use (Mean ± SD)

Licensed Assistant

Resident's status

82)

186)

Confused

3.3

1.0

2.6

1.2*

Physically harms self/others

2.1

.87

1.9

.86

Pulls out IV/NG tube

2.0

.68

1.9

.86

Left unattended in toilet

3.4

1.2

2.9

1.2t

Gets out of bed by self

3.5

1.0

3.0

1.2t

Slides out of chair

2.4

1.1

2.1

.92*

Falls frequently

2.2

1.0

2.0

.96*

* <.05.
t <.01.
t <.001.

§ Lower scores indicate greater likelihood of recommending
restraint use.

564

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

A major finding of the study is the identifi-
cation of factors that contribute to recommenda-
tions of restraint use. The most significant factor
appears to be the perception that the staff would
be harmed and the safety of residents would be
jeopardized if restraints were reduced; also signif-
icant is that staff working the day shift are less
likely to recommend physical restraints than are
those working the evening and night shifts.

These findings are consistent with the liter-
ature, including the finding that agitated pa-
tients are frequently medicated and physically re-
strained. Evans and Strumpf report that
prevention of injury to self or others is the most
frequently cited rationale for the use of physical
restraints (7). Agitated patients can be abusive to
themselves as well as to others. Investigators
such as Catchen (8) report that most accidents
involve wheelchairs and occur during the day,
whereas the most serious falls occur out of bed at
night. Therefore, it is understandable why the
JHHA staff who work the day shift — when more
staff members are present — would be less likely
to recommend restraints; serious falls are less
likely to occur on their shift.

The finding that licensed staff are less likely
to recommend restraints suggests that nursing
education might usefully develop different pro-
grams that address the specific needs of both the
licensed staff and nursing assistants. Another
consideration is that nursing assistants have di-
rect contact with patients and may feel more re-
sponsible if an injury occurs once restraints are
removed. They also have lower status than li-
censed staff and may be inhibited or uncomfort-
able about initiating such a change as removing a
restraint. The fact that many were unwilling to
divulge personal information suggests fear of re-
prisal or lack of confidence.

Summary and Conclusions

This study supports the importance of assess-
ing the intrinsic and extrinsic attitudes and per-
ceptions of staff on physical restraints. Any pro-
gram developed to remove physical restraints in
nursing homes should focus on safety issues and
behavioral interventions that prevent injury to
staff and resident. Environmental adaptations
that would enhance residents' safety, such as
alarms and wedge cushions, would be useful in a
restraint reduction program. Educating staff
about the risks of physical restraints and about
safe alternative interventions might change their
attitudes. By developing an educational program
that focuses on a few residents at a time and by
involving a multidisciplinary team, a truly indi-
vidualized approach to restraint-free care for res-
idents might be developed.

References

1. Williams CC. The experience of long-term care in the fu-

ture. J Gerontol Soc Work 1989; 14(l/2):3-18.

2. Evans LK, Strumpf NE. Patterns of restraint: a cross-

cultural view. Gerontologist 1987; 27:124A.

3. Department of Health and Human Services, Medicare,

and Medicaid. Requirements for long-term care facili-
ties. Federal Register 1989; 54(February 2):5316.

4. Miller M. Iatrogenic and nurisgenic effects of prolonged

immobilization of the aged. J Am Geriatr Soc 1975;
23(8):360-369.

5. Strumpf NE, Evans LK. Physical restraints of the hospi-

talized elderly: perceptions of patients and nurses. Nurs
Res 1988; 37(3):132-137.

6. McHutchion E, Morse JM. Releasing restraints — a nurs-

ing dilemma. J Gerontol Nurs 1989; 15(2): 16-21.

7. Evans LK, Strumpf NE. Tying down the elderly: a review

of the literature on physical restraints. J Am Geriatr
Soc 1989; 37(l):65-74.

8. Catchen H. Repeaters: inpatient accidents among the hos-

pitalized elderly. Gerontologist 1983; 23(3):273-276.

Research Activities in the Department of
Geriatrics and Adult Development

John H. Morrison, Ph.D., Jon Gordon, M.D., Ph.D., Myron Miller, M.D., Howard Fillit, M.D., and

Diane E. Meier, M.D.

Since its establishment ten years ago, the De-
partment of Geriatrics and Adult Development at
Mount Sinai Medical Center has always had a
major commitment to its research programs.
These programs cover a wide spectrum, from the
molecular biology and neurobiology of aging to
age-related changes in bone homeostasis in hu-
mans. The research activities of the department
include laboratories housed entirely within the
department and laboratories established through
liaisons with other departments at Mount Sinai
Medical School interested in the biology of aging.
The activities and interests of the five major re-
search programs within the department are sum-
marized below.

Molecular Genetics of
Neurodegeneration: Transgenic
Studies of Superoxide Dismutase

Jon Gordon, M.D., Ph.D.

One of the few surviving theories of aging is
the oxygen-free-radical theory, which postulates
that aging is a manifestation of cumulative oxy-
gen-free-radical damage to macromolecules. Ex-
amples of such damage include lipid peroxidation
and mutations of DNA. Although it is unlikely
that this or any other single hypothesis will ex-
plain all aspects of aging, evidence is accumulat-

From the Department of Geriatrics and Adult Development,
The Mount Sinai Medical Center, New York City. Address
reprint requests to John H. Morrison, Ph.D., Professor and
Co-Director, Fishberg Research Center for Neurobiology and
Willard T. C. Johnson Research Professor of Geriatrics and
Adult Development, One Gustave L. Levy Place, Box 1065,
New York, NY 10029.

ing that oxygen-free radicals may play an impor-
tant role in the decline of function in the central
nervous system (CNS): Humans or experimental
animals with altered expression of Cu/Zn super-
oxide dismutase (SOD-1), a key enzyme in the
oxygen-free radical scavenging system, exhibit
either accelerated or delayed degenerative
changes in the CNS. For example, transgenic
mice with increased expression of SOD-1 are re-
sistant to drug-induced neurodegeneration, and
humans carrying a variety of point mutations in
the SOD-1 gene are subject to familial, autosomal
dominant amyotrophic lateral sclerosis (FALS),
an age-related degenerative disease that mainly
involves motor neurons.

One of my major research objectives is to
clarify the relationship between oxygen-free rad-
icals and aging in the CNS. My laboratory has
cloned the mouse genomic sequence encoding
SOD-1 (1), and was also the first laboratory to
produce a transgenic mouse (2). I and my col-
leagues have now produced transgenic mice car-
rying additional copies of the mouse SOD-1 gene.
These animals are currently being studied for ex-
pression of the foreign gene, and will ultimately
be examined for lifespan and resistance to dam-
aging effects of oxygen-free radicals.

In addition, my group has created mutant
SOD-1 genes that correspond to the mutant genes
found in humans with FALS. These modified
genes will be microinjected into pronuclear stage
embryos (2), and resultant transgenic mice will
be studied for neurodegenerative disorders.

Finally, I am employing targeted mutagene-
sis in embryonic stem (ES) cells to destroy the
SOD locus. With the ES cell system, it is possible
to substitute mutated, inactive genes for endoge-
nous mouse loci by homologous recombination.

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

565

566

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

EMBRYONAL STEM CELL - MEDIATED GENE TRANSFER

DMGT

IMPLANTATION ^
CHIMERIC MOUSE

Injection of transfected
embryonal stem cell into
normal mouse blastocyst

79'

NORMAL MOUSE

50% of progeny
carry new gene

Fig. 1. Production of transgenic mice by use of embryonic
(embryonal) stem (ES) cells. ES cells are maintained in cul-
ture and subjected to DNA-mediated gene transfer (DMGT,
top), for insertion of genes or gene substitution by homologous
recombination. Transformed cells are then injected into the
cavity of a normal mouse blastocyst and implanted. The re-
sulting mouse is a genetic chimera that often has a striped
coat (center). If these chimeras, usually males, produce sperm
derived from the ES cells, an Fj hybrid mouse carrying the
genetic modification in all cells can be produced (bottom). This
animal can then be bred to establish a new transgenic strain.

The ES cells with such genetic modifications can
then be inserted into the cavity of a normal mouse
blastocyst, whereupon they can be induced to re-
sume normal development and give rise to nor-
mal tissues of adult, genetically mosaic mice. If
ES cell derivatives colonize the germ cell popula-
tion of a chimera, the genetically engineered mu-
tation can be transmitted to progeny to establish
mutant strains of animals. The scheme for con-
struction of transgenic mice by the ES cell ap-
proach is outlined in Fig. 1.

Animals carrying mutated, functionally in-
active SOD-1 genes should provide important
new insights into the role of oxygen free radicals
in aging. Further experiments include use of mi-
croinjection and homologous recombination to re-
place the normal murine SOD-1 locus with SOD-1
genes that have been modified in vitro to carry
mutations specific for FALS in humans. Studies

of transgenic mice constructed carrying the al-
tered genes should help clarify the relationship
between altered SOD-1 activity and neurodegen-
erative disorders such as ALS.

My laboratory is also using transgenic tech-
nology to determine whether modifications of
SOD-1 expression actually affect the mutation
rate in vivo. For these experiments, my labora-
tory colleagues are inserting genes that can be
recovered from the animals by molecular cloning
and conveniently screened for the presence of mu-
tations. These "mutation monitor genes" will be
inserted into the strains with modified SOD-1
genes by breeding, and then animals with both
genes will be studies for mutations accumulation.

Age-Related Changes in
Neuroendocrine Function

Myron Miller, M.D.

Overproduction of the posterior pituitary
hormone vasopressin occurs during aging in hu-
mans and in animals as part of the normal aging
process and as a result of a number of diseases (3,
4). Thus, the development of animal models of
increased vasopressin production can allow stud-
ies to be carried out which will help clarify the
impact of increased hormone secretion on a vari-
ety of systems during the course of aging. One
focus of research has been studies of the physiol-
ogy and pathophysiology of fluid and electrolyte
disturbances, which are common in the aged and
may have significant clinical consequences, in-
cluding alterations in central nervous system
function.

Through application of transgenic technol-
ogy, we in my laboratory have created a trans-
genic mouse model of chronic vasopressin hyper-
secretion (5-7). In this animal, the rat gene for
vasopressin has been incorporated into the mouse
genome. Data obtained clearly demonstrate that
the transgenic rat gene functions in the mouse
and that the transgene is capable of producing
vasopressin in tissue sites where the hormone is
normally expressed (5, 6). Thus, it has been dem-
onstrated that increased vasopressin synthesis
takes place in the hypothalamus of the mouse as
well as in the cerebral cortex (Fig. 2). As a con-
sequence of transgene function, increased
amounts of vasopressin are secreted into the pe-
ripheral circulation and are excreted into the
urine. Data demonstrate that the level of overpro-
duction of the hormone is approximately three to
five times that which would occur in a normal
mouse (Fig, 3).

Vol. 60 No. 6

RESEARCH— MORRISON ET AL.

567

Physiological studies in the transgenic ani-
mals have demonstrated that the incorporated
transgene responds in a normal fashion to chal-
lenges which lead either to stimulation or sup-
pression of hormone production. There is some ev-
idence that chronic exposure to increased
amounts of vasopressin results in activation of
compensatory systems for body water regulation
so that chronic hormone overproduction does not
lead to fluid overload. In addition, there is evi-
dence suggesting that chronic exposure of the
blood vessels to vasopressin leads to a reduction
in blood pressure, again suggesting that compen-
satory mechanisms may develop to offset the
acute effects of the hormone (7).

Previous work from this laboratory has es-
tablished that vasopressin is capable of affecting
central nervous system activity by modifying
brain functions involved in expression of learning
and memory ability (8). We have now conducted
studies in the transgenic mice which demonstrate
that these animals with chronically increased
hormone secretion alter their behavior in the di-
rection of increased arousal and increased ease of
adaptation to new environments (9). These obser-
vations raise the possibility that vasopressin or
its analogs may have therapeutic benefits in
treatment of cognitive disorders.

In addition to the animal studies, clinical re-
search has been conducted on the influence of ag-
ing on hormonal systems involved in water regu-
lation. We have demonstrated that the production
of atrial natriuretic hormone, a substance in-
volved in sodium regulation in the body, is in-
creased during normal aging and that secretion of
this hormone is markedly increased following ex-
ercise. These observations have been utilized in
studies evaluating hormonal function during con-
gestive heart failure to examine the role of al-
tered hormone secretion in the fluid disturbances
which are so prevalent in this disorder.

Other clinical investigations have focused on
the regulation of vasopressin release in individu-
als with various types of dementia. Since cholin-
ergic deficits are characteristic of patients with
Alzheimer's disease and since cholinergic mecha-
nisms are involved in the regulation of vaso-
pressin synthesis and release, there is reason to
suspect that patients with Alzheimer's disease
may have alterations in their ability to regulate
vasopressin. Studies are currently under way to
determine if hormonal responses to cholinergic-
mediated stimuli will be helpful in differentiating
patients with Alzheimer's dementia from those
with dementia of other cause.

NEURAL AVP CONTENT IN HOMOZYGOUS
AND HETEROZYGOUS TRANSGENIC (F-6)
AND CONTROL MICE
HYPOTHALAMUS AVP (mU)

PITUITARY AVP (mU)

CONTROL HETEROZYGOUS HOMOZYGOUS

(5) (6) (5)

Fig. 2. Arginine vasopressin (AVP) content in
the hypothalamus and frontal-temporal cortex is
significantly increased in both heterozygous and
homozygous transgenic mice, being greater in
homozygous than in heterozygous tissues. There
were no differences in pituitary AVP content.

Neuroanatomic and Molecular Determinants
Of Vulnerability in Alzheimer's Disease

John H. Morrison, Ph.D.

Much of the research in my laboratory in-
volves trying to link molecules that have been
implicated in neuronal degeneration to the cells
and circuits that are at risk in Alzheimer's dis-
ease (AD). For example, a wide range of both
qualitative and quantitative studies on the distri-
bution of neuritic plaques (NP), neurofibrillary
tangles (NFT), neuron loss, and synapse loss sug-
gest that there are two classes of cortical circuits
particularly devastated in AD: (a) the corticocor-
tical projections within neocortex that link the
association areas of prefrontal, temporal, pari-
etal, and limbic cortices; and (b) the parahippo-

568

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

PLASMA AND URINE AVP
IN F-6 HOMOZYGOUS AND CONTROL MICE

CONTROL HOMOZYGOUS

(51 <5)

Fig. 3. Both plasma arginine vasopressin
(AVP) concentration and urinary AVP excretion
are significantly increased in transgenic mice ho-
mozygous for the rat AVP gene.

campal and hippocampal circuits that project
from the entorhinal cortex to the hippocampus, as
well as those that provide the primary outflow
from the hippocampus to related cortical areas
(10). Although it is clear that the pyramidal cells
furnishing such projections suffer a devastating
loss in AD, the cellular and molecular basis of
such differential vulnerability in AD or any other
neurodegenerative disease is less clear. I and my
colleagues have hypothesized that neuron classes
and circuits that possess heightened vulnerabil-
ity to the neurodegenerative process underlying
AD are likely to possess a unique neurochemical
profile that is both essential to their particular
role in cortical function, and causally linked to
their vulnerability in AD and other neurodegen-
erative disorders (11). More specifically, the neu-
rochemical and cellular attributes of pyramidal
cells in neocortex that furnish long corticocortical
connections and are presumably crucial for that
particular highly developed, specialized role in
neocortex might also be linked to their vulnera-
bility. Many of this laboratory's current efforts
are directed at developing quantitative neuro-
pathologic and experimental data that test this
hypothesis and reveal the critical cellular and
molecular links to both normal function and vul-
nerability in both neocortex and hippocampus.

Recent studies have demonstrated striking
correlations between the presence of certain pro-
teins, such as neurofilament protein (NFP), and
vulnerability in AD, whereas other proteins, for
instance calcium-binding proteins (CABP), have
been associated with resistance to degeneration.
In the case of NFP, several factors link the pres-
ence of this class of cellular proteins to vulnera-
bility in AD:

First, pyramidal cells that contain particu-
larly high levels of the NFP triplet are highly
prone to degeneration or NFT formation in the
neocortex of AD patients (12-14).

Second, the neurons that appear to exhibit
the highest level of vulnerability in AD, the layer
II neurons of the entorhinal cortex, contain par-
ticularly high levels of the NFP triplet, and NFPs
are present in the transitional forms of the NFT
that can develop in these neurons during normal
aging (14).

Third, transgenic animals that contain the
human gene for the middle molecular weight neu-
rofilament subunit, NF-M, display several reflec-
tions of pathology associated with human neuro-
degenerative disorders, one of which is very similar
to the NFT seen during normal aging and AD (15).

In addition, NFP are selectively present in
certain subgroups of neurons that furnish corti-
cocortical projections across association areas in
monkeys which are also likely to correspond to
the subgroup of projection neurons that are par-
ticularly vulnerable in AD (16) (for example, tem-
poral to prefrontal projection neurons). The
CABPs parvalbumin, calbindin, and calretinin
may play an important role in conferring resis-
tance to degeneration on a particular subset of
neurons. The interneurons containing any one of
these three proteins have been shown to be resis-
tant to degeneration in both AD and ischemia
(see ref. 10 for review), and these proteins may
buffer intracellular calcium, thereby protecting a
neuron from calcium-mediated excitotoxicity.

Recently, laboratory staff members turned
their attention to the glutamate receptors
(GluRs), both for intrinsic interest in respect to
excitatory circuits in cerebral cortex, and as a
third group of proteins that might be a critical
element in a neurochemical profile of vulnerabil-
ity for AD as well as other neurodegenerative dis-
orders, such as ALS or ischemia. Recent develop-
ments in glutamate receptor neurobiology
suggest that while excitatory circuits and gluta-
mate receptors might be ubiquitous in the cere-
bral cortex, clearly the molecular diversity exists
to allow for extraordinary specificity in the
postsynaptic profile of a given excitatory circuit.

Vol. 60 No. 6

RESEARCH— MORRISON ET AL.

569

Thus, a more specific version of the hypothe-
sis stated above could be developed in that par-
ticular GluR subunit combinations and GluR
family interactions might subserve a subset of ex-
citatory circuits in the cerebral cortex, and both
the presence of and putative age-related shifts in
such profiles might be linked to differential vul-
nerability of certain circuits in disease states
such as Alzheimer's disease. Glutamate-receptor-
mediated excitotoxicity has not only been linked
to neurodegenerative disorders as a powerful
mechanism for neuron death, but, more specifi-
cally, has been linked to alterations of the cyto-
skeletal proteins that are implicated in NFT for-
mation. With the use of riboprobes and subunit
specific antibodies, it is now feasible to character-
ize in detail the GluR specificity linked to identi-
fied excitatory cortical circuits, and develop de-
tailed subunit- and family-specific GluR profiles
of both vulnerable and resistant neurons that will
reveal the degree to which certain GluR subunits
exhibit circuit-linked specificity and might be im-
plicated in the differential vulnerability apparent
in neurodegenerative disorders. Thomas Moran of
the Hybridoma Core Facility at Mount Sinai and
I have in collaboration generated several mono-
clonal antibodies to various glutamate receptor
subunit proteins. Double labeling studies have
shown that the neurofilament-containing neu-
rons that are vulnerable to degeneration in AD
also contain the subunits that form kainate and
AMPA receptors (17). The studies with the anti-
body to kainate subunits demonstrated marked
variability in density and distribution of kainate
receptor immunoreactivity across layers and ar-
eas of monkey neocortex.

These results indicate that GluR5/6/7 sub-
unit immunoreactivity appears to be more dense
in higher-order association areas of the frontal
and parietal lobes than in the primary sensory
areas SI and VI (18). These qualitative observa-
tions were confirmed quantitatively by compar-
ing the densities of stained somata in areas 9, VI
and V2. These data raise the possibility that the
functions of higher-order association areas may
be particularly vulnerable to kainate receptor-
mediated toxicity. Similar analyses are now un-
der way for AMPA and NMDA receptor subunit
distribution in neocortex and hippocampus.

Proteoglycans in Vascular and
Other Diseases

Howard Fillit, M.D.
Role in Vascular Disease. Vascular heparan
sulfate proteoglycans (vHSPG) play an important
role in the structure and function of the vascula-

ture, including endothelial cell adhesion and
basement membrane structure, vascular perme-
ability barrier integrity, normal hemostasis, li-
polysis, and vascular repair (19). Vascular
heparan sulfate (HS), a glycosaminoglycan, plays
a major role in the capillary permeability barrier
(20). HS also plays an important role in the deli-
cate balance of coagulation through a variety of
mechanisms which ultimately, at least in part,
inhibit thrombin activity. Endothelial vHSPG
binds antithrombin III (21) and enhances the in-
hibition of thrombin more than a thousandfold
through the formation of a ternary complex.
Thus, heparan sulfate is the native molecule on
the endothelial cell surface that provides the an-
ticoagulant functions associated with heparin.

Numerous studies have demonstrated that
glycosaminoglycans (GAGs) such as hyaluronate,
chondroitin sulfate, heparin, and heparan sulfate
are both immunogenic and immunospecifically
antigenic. Recently, monoclonal antibodies spe-
cific for heparan sulfate (16, 22-24) and heparin
(25), a GAG related to heparan sulfate, have been
reported. Fillit and others have described au-
toantibodies to GAGs which are reactive with
specific antigenic determinants of HS and hepa-
rin (16, 22-27).

Studies of autoimmunity to HS, including
both active (28) and passive transfer (29) immu-
nization models, the presence of the pertinent an-
tibody at the site of injury in animals and humans
with the disease (30, 31), and a correlation of the
antibody with clinical disease activity (31, 32) es-
sentially fulfill the criteria necessary to demon-
strate that autoantibodies to HS GAG are patho-
genic (33). Of interest, heparin treatment of
animals with vascular autoimmunity neutralizes
autoantibodies thought to be responsible for vas-
cular injury (34). Monoclonal antibodies to HS
created in my laboratory from autoimmune mice
inhibited heparan sulfate accelerated formation
of thrombin-antithrombin III complex (23), and
thus may promote a procoagulant state and
thrombosis, suggesting a noninflammatory im-
mune-mediated mechanism of vascular injury in-
duced by autoimmunity to HS. Finally, in
addition to autoimmunity to HS immunodetermi-
nants, antibodies to HSPG protein core immuno-
determinants cause vascular injury in animal
models (35, 36).

Initial studies have focused on vascular au-
toimmunity in systemic lupus erythematosus
(SLE), an intensively investigated autoimmune
disease. These studies have demonstrated au-
toantibodies to intact vHSPG in patients with
systemic lupus erythematosus, including both HS

570

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

and the protein core components (27, 37, 38). Re-
cent data from this laboratory also indicate the
presence of antiheparin antibodies in human vas-
cular disease sera which inhibit formation of TAT
complexes (26).

Autoantibodies to vHSPG were also found in
aged patients with various forms of vascular dis-
ease, including stroke and dementia (38, 39).
Walford (40) initially proposed that autoimmu-
nity may play a role in vascular disease during
aging. Vascular disease in old age is frequently
immunologically mediated (41). In fact, giant cell
arteritis, a common immunologic vascular dis-
ease of the aged, is associated with degenerative
vascular disease (42), and immune mechanisms
may cause atherosclerosis (43). Thus, my col-
leagues and I are continuing to investigate the
role of autoimmunity to vHSPG as a cause of vas-
cular disease in the aged.

Role in Amyloidosis and Alzheimer's Dis-
ease. Alzheimer's disease (AD) is a neurodegen-
erative disorder characterized by a loss of large
neurons and the extracellular accumulation of
amyloid deposits containing a peptide (the A4
peptide) which is derived from a larger amyloid
precursor protein (APP). APP probably plays a
role in the pathogenesis of AD, either as a pri-
mary event or as part of the evolving pathologic
process which leads to neuronal loss and demen-
tia (44).

Proteoglycans also play an important role in
the brain, including CNS development, neurite
outgrowth, and neuronal cell adhesion (45).
HSPGs codeposit concurrently with amyloid pro-
teins in senile plaques in AD and other diseases
and are thought to play a fundamental role in all
forms of amyloidosis (46). Using monoclonal an-
tibodies (mAbs) to vHSPG protein core, the pres-
ence of HSPG in senile plaques and in neurons in
AD has been demonstrated (47, 48). High affinity
binding of different forms of APP to HSPG (49,
50), high affinity binding of both vascular and
neuronal HSPG to APP (45, 51), and high affinity
binding of HSPG to the A4 peptide actually de-
posited in senile plaques have all been demon-
strated.

Essentially four hypotheses have been put
forward as pathological mechanisms in which
HSPG may contribute to amyloidosis. These in-
clude (a) initiation of fibrillogenesis, (b) enhance-
ment of fibril formation, and (c) decreased amy-
loid distribution. All of these hypotheses flow
from an initial binding interaction between APP/
A4 and HSPG. Since APP/A4-HSPG binding ap-
pears to be physiologic, abnormalities in APP/A4-

HSPG binding may account for the pathological
mechanisms cited.

Finally, using monoclonal antibodies to
vHSPG as markers of vascular pathology, H. Fil-
lit has investigated the presence of microvascular
alterations in Alzheimer's disease. Recent studies
have demonstrated statistically significant de-
creases in microvascular density in regions of the
brain affected by AD (52). It is not currently
known if these microvascular changes are a pri-
mary or a secondary event in the disease process,
but in either case could be contributing to neuro-
nal death via the resultant disorder in vascular
supply.

Radical Differences in
Bone Homeostasis

Diane E. Meier, M.D.

Available data suggest a significantly lower
incidence of fracture and higher bone mass in
women of African (black) ancestry than in women
of European (white) ancestry. Since adult bone
loss occurs universally as a function of age, rea-
sons for the racial difference in bone mass and
fracture risk are not immediately apparent. Bone
mineral content, which correlates with bone
strength and fracture risk, is a function of at least
two factors: peak bone mass achieved at skeletal
maturity and subsequent rates and duration of
bone loss. The lower fracture incidence among
black women may be due to (a) higher peak bone
mass at skeletal maturity, so that despite compa-
rable age-related bone loss, blacks reach the frac-
ture threshold less frequently than whites; (b)
age-related bone loss that begins later, is less se-
vere, or occurs in different skeletal sites; or (c) no
difference in bone mass and rate of loss in black
and white women, but differences in other risk
factors, such as frequency of falls.

This study of bone homeostasis in black and
white women was initiated with two hypotheses:
(a) the racial difference in fracture prevalence is
due to detectable racial differences in adult bone
homeostasis; and (b) comparison of groups with
high and low fracture risk can provide useful in-
sights into the etiology of osteoporosis and pro-
vide directions for future research.

Thus, the research was designed to assess the
natural history and the determinants of bone loss
in a longitudinal study of healthy white and black
pre- and post-menopausal women.

Cross-sectional data from this study showed
statistically significantly higher radial and ver-
tebral bone mass in healthy black women, both

Vol. 60 No. 6

RESEARCH— MORRISON ET AL.

571

premenopausal and postmenopausal (53). How-
ever, there was substantial overlap in bone den-
sity between the two races, more than 25% of
healthy black women having vertebral bone den-
sity below the mean for white women. Further-
more, 5% of the black and 17% of the white
women were noted to be below the theoretical
fracture threshold, suggesting that there may be
a previously unrecognized subgroup of otherwise
healthy black women who are at increased risk
for osteoporotic fracture. Bone density at both ra-
dial and vertebral sites declined significantly as a
function of age in both white and black women by
cross-sectional analysis.

Recently analyzed longitudinal data from
this study reveal statistically significant pre-
menopausal bone loss in both white and black
women, a finding which has major implications
for the prevention of osteoporosis in an age group
previously neglected in epidemiologic prevention
studies.

Although both white and black postmeno-
pausal women also demonstrated statistically sig-
nificant bone loss, among women less than 5
years postmenopausal a significant racial differ-
ence in radial (but not spinal) bone loss was ob-
served; radial bone density declined by 1.1% per
year in the blacks, versus 2.2% per year in the
whites (p = .05). This racial differential in early
menopausal bone loss is a new observation, and
may be of important etiologic significance in the
racial disparity in fracture risk.

Several biochemical measures of bone turn-
over were measured at baseline, including fasting
and 24-hour urinary calcium excretion, urinary
hydroxyproline, serum alkaline phosphatase, and
osteocalcin. Interestingly, in both pre- and post-
menopausal women, measures of bone turnover
were significantly lower statistically in black
than in white women (54). These findings may be
reflected in the slower early menopausal bone
loss observed in the black subjects of this study.

To evaluate the vitamin D-parathyroid hor-
mone endocrine system and its relationship to
racial differences in bone density, serum concen-
trations of these hormones were measured. Inter-
estingly, and in contrast to previously published
results, in these healthy, normal weight, well-
nourished middle-class white and black women
there were no major differences in 25 hydroxyvi-
tamin D concentration, nor was there any evi-
dence of secondary hyperparathyroidism in the
blacks; previous studies have shown low levels of
25 hydroxyvitamin D and associated secondary
hyperparathroidism in poor, obese black subjects

(55). Overall, these data suggest that the vitamin
D-parathyroid hormone endocrine system is not
the major etiologic factor contributing to the ra-
cial differences in bone density. This finding un-
derscores the importance of conducting racial
comparisons in groups which are as comparable
as possible for every controllable parameter with
the exception of race. In the past, differences
which were probably due to socioeconomic status
and malnutrition were inappropriately attributed
to racial differences.

To assess the impact of exercise on bone den-
sity, assessment of habitual physical activity by
questionnaire was obtained, revealing a signifi-
cant correlation between lumbar bone density
and average daily weight-bearing activity in both
white and black women. However, the racial dif-
ferences in bone density observed in these sub-
jects could not be statistically accounted for by
differences in physical activity.

No racial differences in gonadal and adrenal
steroids were observed, and steroid hormone con-
centration did not correlate with bone density of
either the radius or the spine, suggesting that the
racial differences in bone density observed in this
study are not attributable to racial differences in
either adrenal or follicular phase gonadal hor-
mone concentration.

These findings to date support the initial hy-
potheses and demonstrate, for the first time,
striking evidence of racial differences in bone ho-
meostasis and calcium metabolism in pre- and
post-menopausal women. In addition, the obser-
vation of significant premenopausal bone loss in
both racial groups has important implications for
the initiation of new research and preventive tri-
als which should focus on intervention during
childhood, adolescence, and early adulthood.

References

1. Benedetto MT, Anzai Y, Gordon JW. Isolation and anal-

ysis of the mouse genomic sequence encoding Cu/Zn su-
peroxide dismutase. Gene 1991; 99:191-195.

2. Gordon JW, Scangos GA, Plotkin DJ, Barbosa JA, Ruddle

FH. Genetic transformation of mouse embryos by mi-
croinjection of purified DNA. Proc Natl Acad Sci USA
1980; 77:7380-7384.

3. Miller M. Influence of aging on vasopressin secretion and

water regulation. In: Schrier RW, ed. Vasopressin. New
York: Raven Press, 1985, 249-258.

4. Miller M, Gold GC, Friedlander DA. Physiological

changes of aging affecting salt and water balance. Rev
Clin Gerontol 1991; 1:215-230.

5. Miller M, Kawabata S, Wiltshire-Clement M, Reventos J,

Gordon JW. Increased vasopressin secretion and re-
lease in mice transgenic for the rat arginine vaso-
pressin gene. Neuroendocrinology 1993; 57:621-625.

572

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

6. Grant FD, Reventos J, Gordon JW, Kawabata S, Miller M,

Majzoub J A. Expression of the rat arginine vasopressin
gene in transgenic mice. Mol Endocrinol 1993; 669:640-
642.

7. Miller M. Chronically increased secretion of vasopressin

in transgenic mice. Ann NY Acad Sci 1993; 669:640-
642.

8. Williams AR, Carey RJ, Miller M. Altered emotionality of

the vasopressin-deficient Brattleboro rat. Peptides
1985; 6(Suppl l):69-76.

9. Miller M, Haroutunian V, Wiltshire-Clement M, Kawa-

bata S. Behavioral and cognitive changes in vaso-
pressin-secreting transgenic mice. Program of the 75th
Annual Meeting of the Endocrine Society 1993; 440.

10. Hof PR, Morrison JH. The cellular basis of cortical discon-

nection in Alzheimer's disease and related dementing
conditions. In: Terry R, Katzman R, Hick K, eds. Alz-
heimer's disease. New York: Raven Press (in press).

11. Morrison JH. Differential vulnerability, connectivity, and

cell typology. Neurobiol Aging, 1993; 14:51-54.

12. Hof PR, Cox K, Morrison JH. Quantitative analysis of a

vulnerable subset of pyramidal neurons in Alzheimer's
disease: 1. Superior frontal and inferior temporal cor-
tex. J Comp Neurol 1990; 301:44-54.

13. Hof PR, Morrison JH. Quantitative analysis of a vulner-

able subset of pyramidal neurons in Alzheimer's dis-
ease: II. Primary and secondary visual cortex. J Comp
Neurol 1990; 301:55-64.

14. Vickers JC, Delacourte A, Morrison JH. Progressive

transformation of the cytoskeleton associated with nor-
mal aging and Alzheimer's disease. Brain Res 1992;
594:273-278.

15. Vickers JC, Morrison JH, Friedrich VL, Elder GA, Perl

DP, Katz RN, Lazzarini RA. Age-associated neurofibril-
lary pathology in transgenic mice expressing the hu-
man mid-sized neurofilament subunit. J Neurosci (sub-
mitted).

16. Kure S, Yoshie O. A syngeneic monoclonal antibody to

murine Meth-A sarcoma (HepSS-1) recognizes heparan
sulfate glycosaminoglycan (HS-GAG): cell density and
transformation dependent alteration in cell surface HS-
GAG defined by HepSS-1. J Immunol 1986; 137:3900-
3908.

17. Vickers JC, Huntley GW, Edwards AM, Moran T, Rogers

SW, Heinemann SF, Morrison JH. Quantitative local-
ization of AMPA/kainate and kainate glutamate recep-
tor subunit immunoreactivity in neurochemically iden-
tified subpopulations of neurons in the prefrontal cortex
of the macaque monkey. J Neurosci 1993; 13:2982-
2992.

18. Huntley GW, Rogers SW, Moran T, et al. Selective distri-

bution of kainate receptor subunit immunoreactivity in
monkey neocortex revealed by a monoclonal antibody
which recognizes glutamate receptor subunits GluR5/6/
7. J Neurosci 1993; 13:2965-2981.

19. Wight TN. Cell biology of arterial proteoglycans. Arterio-

sclerosis 1989; 9:1-20.

20. Cotran RS, Rennke HG. Anionic sites and the mecha-

nisms of proteinuria. N Engl J Med 1983; 309:1050-
1052.

21. Mertens G, Cassiman J-J, Van den Berghe H, Vermylen

J, David G. Cell surface heparan sulfate proteoglycans
from vascular endothelial cells: core protein character-
ization and antithrombin III binding properties. J Biol
Chem 1992; 267:20435-20443.

22. Van den Born J, van den Heuvel J, Bakker MAH,

Veerkamp JH, Assmann KJM, Berden JHM. Produc-

tion and characterization of a monoclonal antibody
against human glomerular heparan sulfate. Lab Invest
1991; 65:287-297.

23. Shibata S, Harpel P, Bona C, Fillit H. Monoclonal anti-

bodies to heparan sulfate inhibit the formation of
thrombin-antithrombin III complexes. Clin Immunol
Immunopath 1993; 67:264-272.

24. David G, Bai XM, Van der Schueren B, Cassiman JJ, Van

den Berghe H. Developmental changes in heparan sul-
fate expression: in situ detection with mAbs. J Cell Biol
1992; 119:961-975.

25. Strauss AH, Travassos LR, Takahashi HK. A monoclonal

antibody (ST-1) directed to the native heparin chain.
Anal Biochem 1992; 201:1-8.

26. Shibata S, Sasaki T, Harpel P, Fillit H. Autoantibodies to

vascular heparan sulfate proteoglycan in systemic lu-
pus erythematosus react with endothelial cells and in-
hibit the formation of thrombin-antithrombin III com-
plexes, (in press).

27. Fillit H, Lahita R. Antibodies to vascular heparan sulfate

proteoglycan in patients with systemic lupus erythema-
tosus. Autoimmunity 1991; 9:159-164.

28. Abrass CK, Cohen AH. Characterization of renal injury

initiated by immunization of rats with heparan sulfate.
Am J Pathol 1988; 130:103-111.

29. Van den Born J, van den Heuvel LPWJ, Bakker MAH,

Veerkamp JH, Assmann KJM, Berden JHM. A mono-
clonal antibody against GBM heparan sulfate induces
an acute selective proteinuria in rats. Kidney Int 1992;
41:115-123.

30. Faaber P, Rijke TPM, van de Putte LBA, Capel PJA, Ber-

den JHM. Cross-reactivity of human and murine anti-
DNA antibodies with heparan sulfate: the major gly-
cosoaminoglycan in glomerular basement membrane.
J Clin Invest 1986; 77:1824-1830.

31. Sasaki T, Hatakeyama A, Shibata S, et al. Heterogeneity

of immune complex-derived anti-DNA antibodies asso-
ciated with lupus nephritis. Kidney Int 1991; 39:746-
753.

32. Termaat RM, Brinkman K, Nossent JC, Swaak AJG,

Smeenk RJT, Berden JHM. Anti-heparan sulfate reac-
tivity in sera from patients with systemic lupus erythe-
matosus with renal or non-renal manifestations. Clin
Exp Immunol 1990; 82:268-274.

33. Bona CA. Postulates defining pathogenic autoantibodies

and T cells. Autoimmunity 1991; 10:169-172.

34. Naparstek Y, Ben-Yehuda A, Madaio MP, et al. Heparin

neutralizes nephritogenic autoantibodies in systemic
lupus erythematosus. Arth Rheum 1991; 33:1554—1559.

35. Makino H, Gibbons JT, Reddy MK, Kanwar YS. Nephri-

togenicity of antibodies to proteoglycans of the glomer-
ular basement membrane-I. J Clin Invest 1986; 77:142-
156.

36. Miettinen A, Stow JL, Mentone S, Farquhar MG. Anti-

bodies to basement membrane heparan sulfate proteo-
glycans bind to the laminae rarae of the gomerular
basement membrane (GBM) and induce subepithelial
GBM thickening. J Exp Med 1986; 163:1064-1084.

37. Fillit H, Shibata S, Sasaki T, Spiera H, Kerr LD, Blake M.

Autoantibodies to the protein core of vascular basement
membrane heparan sulfate proteoglycan in systemic lu-
pus erythematosus. Autoimmunity 1993; 14:243-249.

38. Fillit HM, Mulvihill M. Association of autoimmunity to

vascular heparan sulfate proteoglycan and vascular
disease in the aged. Gerontology (in press).

39. Fillit HM, Kemeny E, Luine V, Weksler ME, Zabriskie

JB. Antivascular antibodies in the sera of patients with

Vol. 60 No. 6

RESEARCH— MORRISON ET AL.

573

senile dementia — Alzheimer's type. J Geront 1987; 42:
180-184.

40. Walford, RL. The immunologic theory of aging, Balti-

more: Williams and Wilkins, 1969.

41. Boesen P, Sorensen SF. Giant cell arteritis, temporal ar-

teritis, and polymyalgia rheumatica in a Danish
county: a prospective investigation, 1982-1985. Arth
Rheum 1987; 30:294-299.

42. Machado EBV, Gabriel SE, Beard CM, Michet CJ, O'Fal-

lon WM, Ballard DJ. A population-based case-control
study of temporal arteritis: evidence for an association
between temporal arteritis and degenerative vascular
disease? Int J Epidemiol 1989; 18:836-841.

43. Minick CR, Murphy GE, Campbell WG. Experimental in-

duction of atheroarteriosclerosis by the synergy of al-
lergic injury to arteries and a lipid rich diet. J Exp Med
1966; 124:635-652.

44. Joachim CL, Selkoe DJ. The seminal role of beta-amyloid

in the pathogenesis of Alzheimer's disease. Alz Dis As-
soc Disord 1992; 6:7-34.

45. Buee L, Ding W, Delacourte A, Fillit HM. Binding of se-

creted human neuroblastoma proteoglycans to the Alz-
heimer amyloid A4 peptide. Brain Res 1993; 601:154—
163.

46. Snow AD, Wight TN. Proteoglycans in the pathogenesis of

Alzheimer's disease and other amyloidoses. Neurobiol
Aging 1989; 10:481^97.

47. Perlmutter LS, Chui HC, Daperia D, Athanikar J. Micro-

angiopathy and the colocalization of heparan sulfate
proteoglycan with amyloid in senile plaques of Alzhei-
mer's disease. Brain Res 1990; 508:13-19.

48. Fillit HM, Buee L, Hof PR, Delacourte A, Morrison JH.

Cortical distribution of abnormal microvasculature and
vascular heparan sulfate proteoglycan positive plaques
in Alzheimer's disease. Soc Neurosci Abstr 1991; 17:692.

49. Narindrasorasak S, Lowery D, Gonzalez-DeWhitt P, Poor-

man RA, Greenberg B, Kisilevsky R. High affinity in-
teractions between the Alzheimer's beta amyloid pre-
cursor proteins and the basement membrane form of
heparan sulfate proteoglycan. J Biol Chem 1991; 266:
12878-12883.

50. Snow AD, Kinsella MG, Sekiguchi RT, Nochlin D, Wight

TN. Binding by a high molecualr weight heparan sul-
fate proteoglycan to the extracellular domain of the
beta-amyloid protein of Alzheimer's disease. Soc Neu-
rosci 1991; 17:1106 (abstract).

51. Buee L, Ding W, Anderson JP, et al. Binding of vascular

heparan sulfate proteoglycan to Alzheimer's amyloid
precursor protein is mediated through the N-terminal
region of the A4 peptide. Brain Res 1993 (in press).

52. Buee L, Bouras C, Hof PR, Surini M, Morrison JH, Fillit

HM, Delacourte A. A qualitative and quantitative im-
munohistochemical study of the microvasculature in
aging and Alzheimer's disease. Soc Neurosci Abstr 1993
(in press).

53. Luckey MM, Meier DE, Mandeli J, DaCosta M, Goldsmith

SJ. Axial and appendicular bone density in white and
black women: Evidence of racial differences in pre-
menopausal bone homeostasis. J Clin Endocrinol Metab
1989; 69:762-770.

54. Meier De, Luckey MM, Wallenstein S, Clemens TL, Or-

woll ES, Waslien CI. Calcium, vitamin D, and parathy-
roid hormone status in young white and black women:
association with racial differences in bone mass. J Clin
Endocrinol Metab 1991; 72:703-710.

55. Meier DE, Luckey MM, Wallenstein S, Lapinski RH,

Catherwood B. Racial differences in pre- and postmeno-
pausal bone homeostasis: association with bone density.
J Bone Miner Res 1992; 10:1181-1189.

Public Policy and Geriatrics:

The International Leadership Center on Longevity and Society of the
Department of Geriatrics and Adult Development at Mount Sinai

Medical Center

Malvin Schechter, M.S.

Many major policy issues for contemporary
America and other nations arise from the demo-
graphics of population aging. The increase in ab-
solute and relative numbers of people over age 65
influences society in many ways, with ramifica-
tions in health care, economics, housing, trans-
portation, nutrition, and politics. For all profes-
sionals in health care, an understanding of
population aging and its institutional ramifica-
tions appear essential to effective practice appro-
priate to all stages of the longer average span of life.

For this reason, the Department of Geriatrics
and Adult Development at Mount Sinai Medical
Center has established a policy center to collect,
analyze, and interpret information and trends for
health-care professionals and for the makers of
public policy. The policy center also serves as a
vehicle for teaching health-care trainees and for
the generation of new information needed to
guide policy development.

This center — the International Leadership
Center on Longevity and Society (U.S.), or
ILCLS — takes the view that the older population
of our country, about 33 million persons, or one in
seven Americans, is a resource for the center it-
self and for society. Contrary to myth, this popu-
lation is mostly well and able. Some older people,
chiefly among the two million persons aged 80
and older, need support from health and social
and other services in order to survive and func-
tion at a reasonable level of satisfaction with life.

From the Department of Geriatrics and Adult Development
and the International Leadership Center on Longevity and
Society (U.S.), The Mount Sinai Medical Center, New York,
NY. Address reprint requests to Malvin Schechter, Associate
Director of the International Leadership Center, Box 1070,
The Mount Sinai Medical Center, New York, NY 10029.

574

The center's primary foci are on policy issues af-
fecting geriatrics and productive aging.

Preserving or developing the conditions un-
der which late life may be productive and secure
with and without infirmity is a social, family, and
personal responsibility. Unfortunately, policies to
promote goals of productive and secure aging are
obstructed by prejudice against older people, or
ageism, and by gaps in information. Often, the
negative image older people have had of them-
selves and the negative image others have held of
older people has restricted the contributions they
could make to society, community, family, and
self Ageism — recognized as a force in political,
economic, and health affairs — is a professional as
well as a general issue.

The center was intended from the beginning
to serve internal as well as external purposes,
that is, to serve as an educational vehicle for fac-
ulty and trainees of the department on public and
private sector trends affecting geriatrics, and to
present our data, analyses, and interpretations to
professional and lay audiences. The goals of the
center are to:

• Identify the salient problems and opportunities
presented by population aging and increased life
expectancies in the United States and elsewhere

• Promote biomedical and socioeconomic re-
search on longevity issues, including productiv-
ity, health maintenance, health and social ser-
vices, quality of life, and social roles

• Stimulate the development of informed policy-
making related to longevity through collabora-
tion among researchers and policymakers in
the public and private sectors

• Disseminate information to the general public,
lay and professional leaders, and the Mount
Sinai community of health-care professionals

The Mount Sinai Journal of Medicine Vol. 60 No. 6 November 1993

Vol. 60 No. 6

PUBLIC POLICY— SCHECHTER

575

Educational Support Functions

Efforts are made to inform medical students,
fellows, and practicing physicians about the pub-
lic and private programs that pay for and other-
wise influence the delivery of services to older
people. The syllabus for the mandatory rotation of
fourth-year medical students through the depart-
ment contains introductory material on the de-
mography and socioeconomics of older Americans
(including utilization of health and social ser-
vices), and on the rationale, history, politics, and
operations of Medicare, Medicaid, and private in-
surance (including how physicians and hospitals
are paid).

In the center's view, the professional and fi-
nancial environments for geriatrics have to
change if geriatrics is to thrive and influence
mainstream medicine. For example. Medicare
was designed to extend to retirees the kind of
health insurance benefits purchased through em-
ployment for working people. This approach ex-
cluded or skimped on long-term or "custodial"
care, preventive services, outpatient drugs, and
mental health services. Physicians were gener-
ously rewarded for procedures associated with
acute care but not at all adequately for time spent
in studies of the medical status of frail patients.
Payment arrangements did not extend beyond
the physician to cover the multidisciplinary geri-
atrics team (including nurse, social worker, and
others). Consequently, the curious result of an in-
surance program designed specifically for older
people was the only weak support to geriatrics.

In 1986, the department held a federally sup-
ported policy conference at Mount Sinai, "Paying
Physicians for Geriatric Care." The conference
utilized a study conducted among experts in geri-
atrics and among rank-and-file practitioners (1,
lA). These studies and the conference, which
were funded in part by the National Center for
Health Services Research and Health Care Tech-
nology, were precursors of the center's growing
interest in health economics as a basis for policy
development.

When the U.S. Congress created the Physi-
cian Payment Review Commission in 1986 to pro-
vide advice on Medicare policy, Robert N. Butler,
the department chairman, was appointed to a
two-year term and gave the commission a focus
on geriatrics issues, using data gathered by the
department at Mount Sinai. The commission rec-
ognized that physicians systematically received
less payment for evaluation and management
services in relation to physician time and effort
than for invasive and imaging procedures (2). Al-
though the commission promoted the develop-

ment of a resource-based relative value scale for
Medicare, there were shortcomings for geriatrics
in the initial approach (for example: case vi-
gnettes for analyzing resource inputs did not in-
clude old or very old patients).

In the area of long-term care, the center's ac-
tivities include papers, editorials, a conference,
and a book. In 1983, a conference, "Long-term
Care Insurance: If Not Now, When?" was held at
Mount Sinai. A paper on health and long-term
care as an issue important to women was pre-
sented in 1986 to a meeting sponsored by Re-
sources for Midlife and Older Women, Inc., and
the American Association of Retired Persons (3).
In 1987, the center produced articles for two ma-
jor medical publications (4, 4 A) and a lecture for
New York State policymakers (5).

Among other publications, the center fur-
nished a section on health insurance for the first
edition (1990) and the second edition (in press) of
The Merck Manual of Geriatrics (6). The Twenti-
eth Century Fund commissioned a study, soon to
be published, on the organization and financing of
a geriatrics-based expansion of Medicare as part
of the general reform of American health care (7).
Informal presentations were made by Robert But-
ler to the Pepper commission (U.S. Bipartisan
Commission on Comprehensive Health Care) in
Washington in 1989 on the design of a public
long-term care strategy and the need for public
support of geriatrics-related research.

On the theme of productive aging, the center
has developed a newsletter called Productive Ag-
ing News which it has been publishing since
1986, experimented with seminars for human re-
sources executives, and held preretirement ses-
sions for employee groups, focusing on health pro-
motion and disease prevention.

Role of the Center in
International Collaboration

The center has sponsored two international
conferences that eventuated in books (8, 8A). Be-
cause of Japanese interest in the issues of produc-
tive aging and long-term care, the center in 1990
assisted in the creation of a counterpart in Tokyo.
Funded by business organizations, the Tokyo cen-
ter is directed by Hideo Ibe, Ph.D., a leader in
social welfare policy. The Tokyo center repub-
lishes Productive Aging News in Japanese and
sponsors its own magazine on productive aging;
the U.S. center has in turn published a digest in
English of Japan's 10-year plan to promote health
and social services for the aged (9) and has also
stimulated a report on older people living alone in
Japan (10). This report is a counterpart to studies

576

THE MOUNT SINAI JOURNAL OF MEDICINE

November 1993

by the Commonwealth Fund Commission on el-
derly people living alone, which was chaired by
Robert Butler during its five-year existence in
1985-1990.

The center is preparing a comprehensive
background paper on optimal environments for
older persons. The paper is to be the basis of dis-
cussions at a Tokyo symposium to be sponsored by
the Sasakawa Health Science Foundation in 1994
(11).

With advice from its Projects Advisory
Group, representing academic, business, and
other interests, the center has adopted a work
plan with these international features:

• Completion of the Almanac on Longevity and
Society: Japan and the United States Compared,
by Charlotte Muller of ILC (U.S.) aided by In
Huh, with the cooperation of ILC (Japan)

• Development of an international scholarly ex-
change program

• Initiation of an international program in com-
parative health economics

• Development of a network of contacts in policy
and technical affairs in the United States and
abroad

Although the center has been known internation-
ally only a short time, it has received a wide
range of visitors. As a quid pro quo for providing
visitors with information on policies and pro-
grams relevant to productive aging and compre-
hensive care of the elderly, the center routinely
and systematically solicits similar information
from its visitors and recruits them into its net-
work of contacts. An interview guide has been
designed for this purpose.

The Role of the
Center Domestically

The center was a cosponsor in May 1993 of a
meeting of leading scientists on strategies to de-
lay the onset of dysfunctions of later life. This was
the first of a series of conferences to which busi-
ness as well as science leaders are invited in an
effort to foster collaboration in scientific develop-
ment.

Another effort to promote collaboration was a
series of center-organized seminars for invited
scholars within the Graduate Center, City Uni-
versity of New York on productive aging and
long-term care. The seminars drew specialists in
such diverse areas as political science, psychol-
ogy, economics, history, housing and environmen-
tal studies, and anthropology.

The center has also developed collaborative

relationships through its Projects Advisory
Group, the Alliance for Aging Research (Wash-
ington), the Third Age Center at Fordham Uni-
versity (New York), the University of Michigan,
the Maxwell School of Citizenship and Public Af-
fairs of Syracuse University, and the University
of California (San Francisco).

The center arranged a summer program to
orient long-term-care policy interns of the Na-
tional Academy of Social Insurance in Washing-
ton, D.C. The center also is host to science writers
of the Columbia University Graduate School of
Journalism. In further association with that
school, the center is developing a program to ori-
ent journalists on scientific and social policy is-
sues of population aging.

In preparation is a book on the impact of lon-
gevity on society, dealing with health care, pen-
sions, and family life. In addition to a factual pre-
sentation of major intersecting social trends
affecting individuals and families, the book will
offer proposals on individual and societal adjust-
ments to improve the quality of life for old and
young, including health and economic security.

Conclusions

Even before the inception of the medical
school 30 years ago. Mount Sinai Medical Center
manifested broad interests beyond clinical medi-
cine in social and public policy issues. The first
dean of the medical school, George James, M.D.,
who was also president of the medical center, had
served as New York City Health Commissioner
and was a leading commentator on questions of
America's adaptation to an older population,
chronic disease and disability, and public health.
He testified before Congress frequently, notably
on issues related to medical education and the
inception of Medicare. The current president of
the medical center, John W. Rowe, M.D., a geri-
atrician, carries on that tradition, as does Robert
Butler, who was the founding director of the Na-
tional Institute on Aging.

The policy center expresses longstanding in-
terests of the Mount Sinai community. As part of
the basic plan of the Department of Geriatrics
and Adult Development, the development of a
policy center was welcomed in 1982 by the Board
of Trustees and Thomas Chalmers, M.D., then
president and dean. Despite difficult times for
fund-raising in the recent past. Mount Sinai itself
has augmented private grants made specifically
to the center.

Because of its milieu in a department and
medical center that daily encounters older people

Vol. 60 No. 6

PUBLIC POLICY— SCHECHTER

577

and their families as well as problems in the de-
livery of services, the policy center is hardly an
ivory tower activity. It has begun to make distinc-
tive contributions at the institutional, national,
and international levels to the definition and
analysis of longevity issues.

References

1. Muller C, Fahs MC, and Schechter M. Primary medical

care for elderly patients: Part 1. Service mix as seen by
an expert panel. J Community Health 1989; 14:2:79-
87.

lA. Fahs MC, Muller C, Schechter M. Primary medical care
for elderly patients: Part 2. Results of a survey of office-
based clinicians. J Community Health, 1990; 14:2:89-
99.

2. Physician Payment Review Commission. Annual Report

to Congress, 1989; Washington DC: The Commission,
1989; page xiii.

3. Schechter M. Long-Term Care and Its Financing: An Is-

sue for Women. Presented at the conference "Women of
Uncertain Age: The Social and Economic Consequences
of Women's Aging," sponsored by N.Y. State Legisla-
tive Committee, the Women's Initiative of the Ameri-
can Association of Retired Persons, and Resources for
Midlife and Older Women, Inc., and held November 14,
1986, in New York City.

4. Schechter M. Editorial: Long-term care insurance. Hosp

Practice 1987;22(5): 11-16.
4A. Schechter M. Book Review: A Will and a Way. JAMA
1986; 256(10): 1365-1366.

5. Schechter M. Issues in health policy for New York State.

Sponsored by the Health Policy and Administration
Consortium of the Capital Area. Albany, NY, Novem-
ber 11, 1987.

6. Schechter M. Health insurance. In: Abrams WB, Berkow

R. eds. The Merck manual of geriatrics. Rahway, NJ:
Merck and Co., 1990.

7. Schechter M. Beyond Medicare: Achieving Long-Term

Care Security. San Francisco: Jossey-Bass Publishing
Co., 1993.

8. Butler RN, Oberlink MR, Schechter M, eds. The promise

of productive aging: from biology to social policy. New
York: Springer Publishing Co., 1990.
8A. Butler RN, Kiikuni K, eds. Who is responsible for my old
age? New York: Springer Publishing Co., 1993.

9. The International Leadership Center on Longevity and

Society (U.S.). The golden plan. New York: The Center,
1990.

10. Sodei T. Elderly people living alone in Japan. New York:

The International Leadership Center on Longevity and
Society (U.S.), 1991.

11. Howe JL, Clark H. The optimal environment for the el-

derly. New York: The International Leadership Center
on Longevity and Society (U.S.) (in press).

List of Journal Reviewers in 1993

Each year the JOURNAL relies on its reviewers as the linchpin in offering its readers
worthwhile, timely, and interesting peer-reviewed articles. The JOURNAL staff
thanks the following reviewers for their time and intellectual effort on manuscripts
submitted to the publication in 1993.

Bernard Baumrin

rvicnara o. riaDer

iviyron iviiiier

Samuel Bloom

Noam Harpaz

Daniel A. Moros

Barbara Brenner

George Haskins

unariotte iviuiier

Howard Bruckner

Tomas Heimann

Burt Myers

HiQwara u . uoiione

0. naroiG neiaerman

Demetrius Pertsemlidis

David J. Bronster

Elizabeth Herries

Hiiiioid. rcayiieia

ijcwis Durrowb

oiid.i(jiii riiracriiiid.n

J-itiWlB Ivclfeind.n

U o n 1 1 1 1 1 ^ o 1

ivdnui tJiriad.1

Rosa.mond Rhodes

J. llLliiiCLO XVCllili

Allan Pi Rn nP'nci1'PiT\

Suzanne Carter Kramer

Andrew S. Kaplan

Jack Rudick

Judy Cohen

Jodi Katz

David B. Sachar

Claudia Colgan

Rhona Keller

Henry Sacks

Charlotte Cunningham-Rundles

Richard Knight

Martin H. Savitz

Kurt W. Deuschle

Isidore Kreel

Harry Schanzer

Thomas Einhorn

Leslie Kuhn

Fenton Schaffner

Joseph Eisenman

Marrick Kukin

David Schonholz

Arthur Figur

Jerome L. Knittle

Gary Slater

Howard Fillit

Richard Lasser

Donald A. Smith

Adrienne M. Fleckman

Neal S. LeLeiko

Alex Stagnaro-Green

Eugene W. Friedman

Charles Lieber

Barry D. Stimmel

Harriet S. Gilbert

Kenneth V. Lieberman

Louis Teichholz

Michael Greenberg

Lynda R. Mandell

Alvin S. Teirstein

Adrian Greenstein

Robert Matz

Carl Teplitz

Donald Gribetz

Charles K. McSherry

Rein Tideiksaar

The Department of Rehabilitation Medicine
The Mount Sinai School of Medicine (CUNY)

The Page and William Black
Post-Graduate School of Medicine

The Rehabilitation Medicine Services
Veterans Affairs Medical Center, The Bronx, NY

sponsor the first international course on

myofascial
PAIN

diagnosis & treatment

emphasizing practical clinical management

course: December 8-11, 1993
workshop: December 12, 1993
The Mount Sinai Medical Center
100th St. & Madison Ave., New York, NY
Stern Auditorium

course director: Andrew A. Fischer, MD, PhD, Chief, Rehabilitation Medicine Service, VAMC, Bronx, NY
faculty: Janet Travail, MD, Hans Kraus, MD, and other leading authorities

Fees: CME credits available

course: $390 physicians designed for pain specialists, including physiatrists,

$290 others orthopedists, anesthesiologists, rheumatologists,

workshop: $120 physicians neurologists, physical and occupational therapists,

$100 others chiropractors, and myotherapists

discounted hotel and travel arrangements for registrants: Zenith Travel (212) 241-9447

for information call The Page and William Black Post-Graduate School of Medicine, Mount Sinai
School of Medicine, Box 1193, One Gustave L. Levy Place, New York, NY 10029

(212) 241-6737

NOVEMBER 1993

MOUNT SINAI SCHOOL OF MEDICINE

FEBRUARY 1994

I Annual Orthopaedic In-Tralning Exam Review

Michael M Lewis, MD

5 Symposium on Venous Diseases

Harry R Schanzer, MD
Robert A. Nabatoff. MD

10-12 Chemotherapy Foundation Symposium XI

Innovative Cancer Chemotherapy for Tomorrow

Holiday Inn Crowne Plaza
Ezra M. Greenspan, MD

I I Evening Seminar: What's New for Cancer
Patients?

Holiday Inn Crowne Plaza
Ezra M Greenspan. MD
Mary-Ellen Siegel, MSW

1 2-1 3 Eastern Group Psychotherapy Society Annual
Conference

Robert Friedman, PhD
Hillel Swiller, MD

12 — 14 Soc for Clinical and Experimental Hypnosis:
Introductory Clinical Hypnosis Workshop

Marianne S Andersen. PhD
Stephen Snyder. MD

17 The Dean's Lecture Series, 4 PU Hatch Aud

Manipulating the Mouse Embryo: From Cells
to Genes

Janet Rossant. PhD, Mt Sinai Hosp of Toronto

DECEMBER 1993

7 Training In Child Abuse Recognition &
Reporting

6 PM Hatch Aud
Katherine Teets Grimm. MD
repeated 2/1, 3/29, 6/28

8 The Dean's Lecture Series, 4 PM Hatch Aud
Using an Ancient Enzyme to Regulate the
Fate of RNA: How Eukaryotes Sense and
Control a Toxic Nutrient

Richard D Klausner. MD, Natl Institutes of Health

8-11 Myofascial Pain: Diagnosis & Treatment

Andrew A. Fischer, MD

JANUARY 1994

(to be Symposium on Women's Health Care Issues
announ- In the 90's: For Primary Care Providers

ced) Michael Brodman, MD
Martin E Goldman, MD
Stacey E Rosen, MD

1 2 The Dean's Lectures Series, 4 PM Hatch Aud

Memory and the Mind

Patricia Goldman-Rakic. PhD
Yale University School of Medicine

1 4 Brachytherapy for Prostate Cancer

Nelson N Stone. MD

19 Queens Hospital Center Internal Medicine

Board Review Course

Wednesday evenings through June 1994
Laguardia Marriott Hotel
Fred Rosner. MD

22 American Academy of Psychiatry and the Law

Conference

Karl M Easton, MD

23- 27 International Trauma Anesthesia & Critical

Care Soc: 1994 Trauma Anesthesia Update

Lake Tahoe. California
Kenneth J. Abrams, MD

24- 28 40th Comprehensive Sinus Surgery Course

William Lawson. MD. DOS
Hugh F, Biller, MD

29-30 Soc for Clinical & Experimental Hypnosis:
Advanced Clinical Hypnosis Workshop

Marianne S Andersen, PhD
Stephen Snyder, MD

The Page and William Black
Post-Graduate School of Medicine
Of Mount Sinai School of Medicine
Of The City University of New York

Continuing
i\/iedicai

Education
Courses

autumn 1993-spring 1994

for information, please call (212) 241-6737

The Page and William Black
Post-Graduate School of the
Mount Sinai School of h/ledicine (CUNY)
is accredited by the Accreditation Council
for Continuing Medical Education (ACCME)
to sponsor continuing medical education
for physicians

unless otherwise indicated, all courses
held on Mount Sinai campus at
100th Street & Madison Avenue, New York City

2-6 Geriatrics Board Review Course

Robert N, Butler. MD. Judith Ahronheim, MD,
Gabriel Gold, MD

5-6 Advanced Hypnosis In Behavioral Medicine,

Psychosomatic Disorders & Consultation Liaise
Psychiatry

Harold J Wain, PhD. Steven Snyder. MD

9 The Dean's Lecture Series, 4 PM Hatch Aud

Molecular Properties of Voltage Gated Ion
Channels

William A Catterall, PhD

Univ of Washington School of Medicine

MARCH 1994

4 Issues in Medical Ethics 1994: Scientific Medici

Alternative Care, and Third-Party Payers

Rosamond Rhodes. PhD

9 The Dean's Lecture Series, 4 PM Hatch Aud

Mechanism & Regulation of Protein Transport
Early in the Secretory Pathway

Randy Schekman, PhD
University of California at Berkeley

5-12 Skull Base Surgery

Chandranath Sen. MD & Peter Catalano, MD

Head & Neck Reconstruction

Mark Urken, MD
Hugh F. Biller, MD

18 25th Annual Communications Disorders

Conference

Arnold Shapiro

APRIL 1994

7-8 Alternate Treatments for Localized Carcinoma
of the Prostate: Radical Perineal Prostatectomy
Brachytherapy, Cryosurgery

Nelson N Stone. MD

13 The Dean's Lecture Series, 4 PM Hatch Aud

New Concepts In Steroid Receptor Activation

Bert O'Malley, MD

Baylor College of Medicine

20-22 Aortic Surgery Symposium IV

New York Hilton & Towers

Randall B Griepp, MD '

22-23 Thyroid Disease and the Eye

Arthur L Mlllman, MD

MAY 1994

7 Conference on Mediastinal Diseases

Steven M Keller, MD
Paul A. Kirschner, MD

8 Echocardiography Workshop

Martin E Goldman, MD

9-13 Consultant's Course In Cardiology

Louis E Teichholz, MD
Richard Gorlin, MD
Simon Dack, MD

1 1 The Dean's Lecture Series, 4 PM Hatch Aud

Signal Transduction by Receptors that
Regulate Tyrosine Phosphorylation

Joseph Schlessinger, PhD
NYU School of Medicine

JUNE 1994

The Dean's Lecture Series, 4 PM Hatch Aud
Receptor Proteln-tyroslne Kinases and
Phosphatases and Their Substrates

Tony Hunter, PhD

The Salk Institute For Biological Studies

Instructions for Authors

THE MOUNT SINAI JOURNAL OF MEDICINE, established in 1934 as The Journal of the Mount Sinai Hospital, is a peer-re-
viewed general medical journal. It is indexed or abstracted in Index Medicus, Current Contents, Science Citation Index, Hospital
Literature Index. International Nursing Index, Excerpta Medica, Chemical Abstracts, Biological Abstracts, and major databases.
Theoretical and basic science articles for the clinician, reviews, reports of clinical or research experience, articles on any subject
affecting the practice of medicine, short notes in any specialty, book reviews, letters, and articles by medical students are
welcome. The Journal is a participant in the Agreement on Uniform Requirements for Manu-scripts Submitted to Biomedical
Journals. The Journal aims at responding to authors of manuscripts within 8 weeks of date of submission.

Authorship Responsibility, Financial Disclosure, and Assign-
ment of Copyright. Each author is required to sign the three
statements on the accompanying page.

Key Words. Supply 3 to 10 key words or phrases at the bottom
of the title page, for use in indexing the article. Use Med-
ical Subject Headings from Index Medicus.

Units of Measurement. Use SI units, giving range or interval
of normal values.

Illustrations. Photographs, line drawings, graphs, and charts
should increase understanding of the text. Line drawings,
graphs, and charts should be professionally prepared. All il-

lustrations should be submitted in triplicate as sharp, high-
contrast glossy prints. Photomicrographs must be 5Vh in.
wide. On the back of each print affix a gum label with the
number of the figure (numbered consecutively in arable nu-
merals as cited in text), title of manuscript, name of senior
author, and the word "top" with an arrow indicating top edge.
Illustrations which are to appear together should be mounted
accordingly as plates and should correspond to each other in
size. Protect the prints by enclosing them in heavy cardboard;
do not use paper clips. Photographs of patients can be pub-
lished only if a copy of the permission is submitted with the
photograph. Eliminate names of patients and hospital
numbers from x-rays.

1. Nadelman RB, Wormser GP. A clinical approach to Lyme
disease. Mt Sinai J Med 1990;57:144-156.

2. International Committee of Medical Journal Editors. Uni-
form requirements for manuscripts submitted to biomedical
journals. N Engl J Med 1991;324:424--128.

When a manuscript is accepted for publication, you will be
asked to confirm the accuracy of all references.

Reprint Orders. A reprint order form is sent to the corre-
sponding author with proofs; return it to the editorial office.

J it„ Li.«««i,"i,faa'stiS^^^^^

THE 3 4805 0369435 9

MOUNT SINAI JOURNAL
OF MEDICINE

Forthcoming:

Grand Rounds

Gene Therapy for Immunodeficiency and Cancer

R. Michael Blaese
National Institutes of Health

Neonatal Herpes Simplex Virus Infections:
Natural History, Pathogenesis, and Preventability

Richard J. Whitley

University of Alabanna School of Medicine

Cytokine Gene Therapy

Bernd Gansbacher

Memorial Sloan-Kettering Cancer Center

The Function of the p53 Tumor Suppressor
Gene and Its Clinical Correlates

Arnold Jay Levine
Princeton University

Mechanisms of Autoantibody Production and
Their Role in Disease

Keith Elkon

The Hospital for Special Surgery

Pathway for Water Absorption in
Isosmotic Transporting Epithelia

Guillermo Whittembury

Institute Venezolano de Investigaciones Cientificas

Theme Issues

JANUARY 1994

Transplantation

Editor: Lewis Burrows

MARCH 1994

Update on Sleep Disorders

Editor: Lee K. Brown

MAY 1994

Cervical Disk Disease:
Controversies in Neurosurgeiy

Editors: Kalmon Post and Martin Savitz

SEPTEMBER 1994

Resuscitation and Airways
Management

Editor: Allan Reed

Available now:

Toward the Conquest of Pain Issues in Perspectives on

Allan p Reed, editor Medicdl Ethics Quality Improvement in

TOlume ^^^1'°' ^' ' ' Daniel A. Moros and Rosamond Rhodes, editors Health Care

^^^^^ volume 60, no. I.January 1993 _ ^ ^ j.

Suzanne Carter Kramer, editor
84 pages »15 volume 60, No. 5, October 1993

1 12 pages *15

Siii>5cr<p«o(u for 1993: $65 indioiduats. US.; $70 all libraries: $75 indioiduals outside US.
To order subscriptions or copies ofbatk issues ($15 each), please send check, payable to The Mount Sinai Journal of Medicine, to the Journal at Box 1094, One GusUoe L Levy Place,
New York, NY 10029-6574; counUr sales at Suite IE, 50 E. 98th Street, New York, NY (phone: 212 241-6108; FAX: 212 722-6386).

Gt L

LtVY

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Texts

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Mount Sinai Journal of Medicine"

See other formats